Eusflat2007 - Fuzzy Logic
european blog for fuzzy logic and technology

Archive for November, 2007

Is One World Government a possibility?

Friday, November 30th, 2007
THERESA asked:



The same financial crisis is this possibility years ago.

The same financial crisis is this possibility years ago argued it was conspiracy.


Josephine

Has anybody read up on the “String Theory?”?

Tuesday, November 27th, 2007
tmajoros@sbcglobal.net asked:



The possibility of at least 11 dimensions existing and if this theory to.


Margaret

What is Ion Cleanse Foot Spa? Understand it Before You Buy

Tuesday, November 27th, 2007
Due to the misleading explanation from other imitation ion cleanse foot spa device claims that the color changes in the water are the indicators that toxin has been pull out from the poles of our foot. This description has cause most of the people believe that color changes indicate the organ has been detoxified.

The fact of the color changing is caused by the chemical reaction of the water, salt, sweat and the electrode. The ion cleanse foot spa device works by conducting a process called ionization. This process is to recharge or energize our organs and tissues in order we will feel fresher and much healthier after that.

A healthy person is supposed to contain 80% of negative electron (ie. Anion) and 20% of positive electron (ie. Cation ) in the body, which is more commonly know as Yin Yang ( +ive & -ive ) balancing.

Ion cleanse foot spa device may normalize the immune system and bring back optimum natural healthy function. The ions created recharge our organs and tissue bringing more ‘chi’ and improve our vital force. Horning Law of Cure-major organ heals first than lesser organ. Healing starts from inside out. Top to the bottom. Majority of the people experience a better well being after using Ion cleanse foot spa.

The color changes alone in water are not a strong yardstick and remain controversial and may only satisfy the layman but not the professional query.

This is the value of this technology and what it can do to facilitate improvement in human health and the immune system.

By using the ion cleanse foot spa device in a long term, it will provides a thorough and efficient way to maintain high energy levels, improves the body’s immune to toxins and maintain long term wellness.

There is evidence of doctors who have observed and reported health improvements on their clients. Many of our customers have given good testimonies to general improvement to well being.

To prove the efficacy or beneficial efficiency of this device, trials have been done by Bob Lim MD (AM) India.BHNS, APHom of Malaysia using E-PLUS ION SPA device on patient through several methodologies. DR.(H) Bob Lim uses the following method to check the efficiency of E-PLUS ION SPA to improve general health condition and any improvement physiologically or pathological change.



Fuzzy logic sphygmomanometer-monitoring systole & diastole blood pressure.



EAV  Morall machine ( German technology) to test major organ energy level and heavy metal contamination ( disease tester).



Pathological laboratory blood & urine test profile.



 

# (1) & (2) were done before and immediately after each session. (3) Periodical monitoring.

Example:

When a handphone’s battery level is low or finish, this will cause us failure to restart the handphone. The only way to restart the handphone and continue to operate is to recharge the battery until a full level.

Same to human body, today we are exposing to the great toxic load and causing different kinds of body sickness. This will causes the battery level in our body decrease and all we need to do is recharge back by using this device in order to increase our body energy.

Go to http://www.aysia.com.my/p-testimonial.php for more information.



By: Ger Hon

About the Author:

We are selling & supplying Ion Cleanse . Welcome to visit us.



Martin

Financial Modelling

Sunday, November 25th, 2007
FINANCIAL MODELING

 

Financial modeling is a process of forecasting performance of a certain asset, using relationships among operating, investing, and financing variables. The central aim of all financial modeling is valuation under uncertainty: how to estimate the value of a security when its future trajectory, or the trajectory of the other securities or economic variables it depends on, is unknown. Usually, financial modeling requires a great deal of spreadsheet work.

 

 

Financial Modeling Application

 

ü       Business valuation, especially discounted cash flow

ü       Cost of capital or WACC

ü       Modeling the term structure of interest rate and credit spread

ü       Option pricing

ü       Real options

ü       Risk modeling

ü       Portfolio problems

 

 

Standard and Premise of Business Value

 

Before the value of a business can be measured, the valuation assignment must specify the reason for and circumstances surrounding the business valuation. These are formally known as the business value    standard and premise of value.

 

Business valuation results can vary considerably depending upon the choice of both the standard and premise of value. For example, a business buyer and seller may bargain to establish the value of business assets that approaches the fair market value standard.

 

However, the value conclusions based on the going concern premise and that of assemblage of business assets may be quite different. One reason is that an operating business creates value by means of its ability to coordinate its capital, human and management resources to produce economic income. The same set of assets not currently used to produce income is generally worth less.

 

Reasons for Business Valuation

 

Business people may need to conduct business valuation for a number of reasons including sale, estate tax planning, estate tax valuation, divorce, business purchase price allocation, collateral documentation, litigation and documenting that a sales price is equitable.

 

Fair market value

 

“Fair market value”, a central standard of measuring business value, is defined as the price at which property would change hands between a willing buyer and a willing seller when the former is not under any compulsion to buy and the latter is not under any compulsion to sell, both parties having reasonable knowledge of relevant facts. See IRS Rev. Rul. 59-60, 1959-1, Cum. Bulletin 237, codified at 26 C.F.R. § 20.2031-1(b).

 

The fair market value standard incorporates certain assumptions, including the assumptions that the hypothetical purchaser is reasonably prudent and rational but is not motivated by any synergistic or strategic influences; that the business will continue as a going concern and not be liquidated; that the hypothetical transaction will be conducted in cash or equivalents; and that the parties are willing and able to consummate the transaction.

 

These assumptions might not, and probably do not, reflect the actual conditions of the market in which the subject business might be sold. However, these conditions are assumed because they yield a uniform standard of value, after applying generally-accepted valuation techniques, which allows meaningful comparison between businesses which are similarly situated.

 

Elements of business valuation

 

Economic conditions

 

A business valuation report generally begins with a description of national, regional and local economic conditions existing as of the valuation date, as well as the conditions of the industry in which the subject business operates. A common source of economic information for the first section of the business valuation report is the Federal Reserve Board’s Beige Book, published quarterly by the Federal Reserve Bank. State governments and industry associations often publish useful statistics describing regional and industry conditions.

 

Financial Analysis

 

The financial statement analysis generally involves common size analysis, ratio analysis (liquidity, turnover, profitability, etc.), trend analysis and industry comparative analysis. This permits the valuation analyst to compare the subject company to other businesses in the same or similar industry, and to discover trends affecting the company and/or the industry over time. By comparing a company’s financial statements in different time periods, the valuation expert can view growth or decline in revenues or expenses, changes in capital structure, or other financial trends. How the subject company compares to the industry will help with the risk assesment and ultimately help determine the discount rate and the selection of market multiples.

 

Normalization of financial statements

 

The most common normalization adjustments fall into the following four categories:

 

Comparability Adjustments. The valuator may adjust the subject company’s financial statements to facilitate a comparison between the subject company and other businesses in the same industry or geographic location. These adjustments are intended to eliminate differences between the way that published industry data is presented and the way that the subject company’s data is presented in its financial statements.

 

Non-operating Adjustments. It is reasonable to assume that if a business were sold in a hypothetical sales transaction (which is the underlying premise of the fair market value standard), the seller would retain any assets which were not related to the production of earnings or price those non-operating assets separately. For this reason, non-operating assets (such as excess cash) are usually eliminated from the balance sheet.

 

Non-recurring Adjustments. The subject company’s financial statements may be affected by events that are not expected to recur, such as the purchase or sale of assets, a lawsuit, or an unusually large revenue or expense. These non-recurring items are adjusted so that the financial statements will better reflect the management’s expectations of future performance.

 

Discretionary Adjustments. The owners of private companies may be paid at variance from the market level of compensation that similar executives in the industry might command. In order to determine fair market value, the owner’s compensation, benefits, perquisites and distributions must be adjusted to industry standards. Similarly, the rent paid by the subject business for the use of property owned by the company’s owners individually may be scrutinized.

 

Income, Asset and Market Approaches

 

Three different approaches are commonly used in business valuation: the income approach, the asset-based approach, and the market approach. Within each of these approaches, there are various techniques for determining the fair market value of a business. Generally, the income approaches determine value by calculating the net present value of the benefit stream generated by the business (discounted cash flow); the asset-based approaches determine value by adding the sum of the parts of the business (net asset value); and the market approaches determine value by comparing the subject company to other companies in the same industry, of the same size, and/or within the same region.

 

In determining which of these approaches to use, the valuation professional must exercise discretion. Each technique has advantages and drawbacks, which must be considered when applying those techniques to a particular subject company. Most treatises and court decisions encourage the valuator to consider more than one technique, which must be reconciled with each other to arrive at a value conclusion. A measure of common sense and a good grasp of mathematics is helpful.

 

Income approaches

 

The income approaches determine fair market value by multiplying the benefit stream generated by the subject company times a discount or capitalization rate. The discount or capitalization rate converts the stream of benefits into present value. There are several different income approaches, including capitalization of earnings or cash flows, discounted future cash flows (“DCF”), and the excess earnings method (which is a hybrid of asset and income approaches). Most of the income approaches consider the subject company’s historical financial data; only the DCF method requires the subject company to provide projected financial data. Most of the income approaches look to the company’s adjusted historical financial data for a single period; only DCF requires data for multiple future periods. The discount or capitalization rate must be matched to the type of benefit stream to which it is applied. The result of a value calculation under the income approach is generally the fair market value of a controlling, marketable interest in the subject company, since the entire benefit stream of the subject company is most often valued, and the capitalization and discount rates are derived from statistics concerning public companies.

 

Discount or capitalization rates

 

A discount or capitalization rate is used to determine the present value of the expected returns of a business. The discount rate and capitalization rate are closely related to each other, but distinguishable. Generally speaking, the discount rate or capitalization rate may be defined as the yield necessary to attract investors to a particular investment, given the risks associated with that investment. The discount rate is applied only to discounted cash flow (DCF) valuations, which are based on projected business data over multiple periods of time. In DCF valuations, a series of projected cash flows is divided by the discount rate to derive the present value of the discounted cash flows. The sum of the discounted cash flows is added to a terminal value, which represents the present value of business cash flows into perpetuity. The sum of the discounted cash flows and the terminal value is the value of the business.

 

On the other hand, a capitalization rate is applied in methods of business valuation that are based on historical business data for a single period of time. The after-tax net cash flow capitalization rate is equal to the discount rate minus the long-term sustainable growth rate. The after-tax net cash flow of a business is divided by the capitalization rate to derive the present value. Capitalization rates may be modified so that they may be applied to after-tax net income or pre-tax cash flows or income. There are several different methods of determining the appropriate discount rates. The discount rate is composed of two elements: (1) the risk-free rate, which is the return that an investor would expect from a secure, practically risk-free investment, such as a government bond; plus (2) a risk premium that compensates an investor for the relative level of risk associated with a particular investment in excess of the risk-free rate. Most importantly, the selected discount or capitalization rate must be consistent with stream of benefits to which it is to be applied.

 

Build-Up Method

 

The Build-Up Method is a widely-recognized method of determining the after-tax net cash flow discount rate, which in turn yields the capitalization rate. The figures used in the Build-Up Method are derived from various sources. This method is called a “build-up” method because it is the sum of risks associated with various classes of assets. It is based on the principle that investors would require a greater return on classes of assets that are more risky. The first element of an Build-Up capitalization rate is the risk-free rate, which is the rate of return for long-term government bonds. Investors who buy large-cap equity stocks, which are inherently more risky than long-term government bonds, require a greater return, so the next element of the Build-Up method is the equity risk premium. In determining a company’s value, the long-horizon equity risk premium is used because the Company’s life is assumed to be infinite. The sum of the risk-free rate and the equity risk premium yields the long-term average market rate of return on large public company stocks.

 

Similarly, investors who invest in small cap stocks, which are riskier than blue-chip stocks, require a greater return, called the “size premium.” Size premium data is generally available from two sources: Morningstars’ (formerly Ibbotson & Associates’) Stocks, Bonds, Bills & Inflation and Duff & Phelps’ Risk Premium Report.

 

By adding the first three elements of a Build-Up discount rate, we can determine the rate of return that investors would require on their investments in small public company stocks. These three elements of the Build-Up discount rate are known collectively as the “systematic risks.”

 

In addition to systematic risks, the discount rate must include “unsystematic risks,” which fall into two categories. One of those categories is the “industry risk premium.” Morningstar’s yearbooks contain empirical data to quantify the risks associated with various industries, grouped by SIC industry code.

 

The other category of unsystematic risk is referred to as “specific company risk.” Historically, no published data has been available to quantify specific company risks. However as of late 2006, new research has been able to quantify, or isolate, this risk for publicly-traded stocks through the use of Total Beta calculations. P. Butler and K. Pinkerton have outlined a procedure using a modified Capital Asset Pricing Model (CAPM) to calculate the company specific risk premium. The model uses an equality between the standard CAPM which relies on the total beta on one side of the equation; and the firm’s beta, size premium and company specific risk premium on the other. The equality is then solved for the company specific risk premium as the only unknown. While this is ground-breaking research, it has yet to be adopted and used by the valuation community at large.

 

It is important to understand why this capitalization rate for small, privately-held companies is significantly higher than the return that an investor might expect to receive from other common types of investments, such as money market accounts, mutual funds, or even real estate. Those investments involve substantially lower levels of risk than an investment in a closely-held company. Depository accounts are insured by the federal government (up to certain limits); mutual funds are composed of publicly-traded stocks, for which risk can be substantially minimized through portfolio diversification; and real estate almost invariably appreciates in value of long time horizons.

 

Closely-held companies, on the other hand, frequently fail for a variety of reasons too numerous to name. Examples of the risk can be witnessed in the storefronts on every Main Street in America. There are no federal guarantees. The risk of investing in a private company cannot be reduced through diversification, and most businesses do not own the type of hard assets that can ensure capital appreciation over time. This is why investors demand a much higher return on their investment in closely-held businesses; such investments are inherently much more risky.

 

Capital Asset Pricing Model (“CAP-M”)

 

The Capital Asset Pricing Model is another method of determining the appropriate discount rate in business valuations. The CAP-M method originated from the Nobel Prize winning studies of Harry Markowitz, James Tobin and William Sharpe. Like the Ibbotson Build-Up method, the CAP-M method derives the discount rate by adding a risk premium to the risk-free rate. In this instance, however, the risk premium is derived by multiplying the equity risk premium times “beta,” which is a measure of stock price volatility. Beta is published by various sources (including Ibbotson Associates, which was used in this valuation) for particular industries and companies. Beta is associated with the systematic risks of an investment.

 

One of the criticisms of the CAP-M method is that beta is derived from the volatility of prices of publicly-traded companies, which are likely to differ from private companies in their capital structures, diversification of products and markets, access to credit markets, size, management depth, and many other respects. Where private companies can be shown to be sufficiently similar to public companies, however, the CAP-M model may be appropriate.

 

Weighted Average Cost of Capital (“WACC”)

 

The weighted average cost of capital is the third major approach to determining a discount rate. The WACC method determines the subject company’s actual cost of capital by calculating the weighted average of the company’s cost of debt and cost of equity. The WACC capitalization rate must be applied to the subject company’s net cash flow to invested equity. One of the problems with this method is that the valuator may elect to calculate WACC according to the subject company’s existing capital structure, the average industry capital structure, or the optimal capital structure. Such discretion detracts from the objectivity of this approach, in the minds of some critics.

 

Once the capitalization or discount rate is determined, it must be applied to an appropriate economic income streams: pretax cash flow, aftertax cash flow, pretax net income, after tax net income, excess earnings, projected cash flow, etc. The result of this formula is the indicated value before discounts. Before moving on to calculate discounts, however, the valuation professional must consider the indicated value under the asset and market approaches.

 

Careful matching of the discount rate to the appropriate measure of economic income is critical to the accuracy of the business valuation results. Net cash flow is a frequent choice in professionally conducted business appraisals. The rationale behind this choice is that this earnings basis corresponds to the equity discount rate derived from the Build-Up or CAP-M models: the returns obtained from investments in publicly traded companies can easily be represented in terms of net cash flows. At the same time, the discount rates are generally also derived from the public capital markets data.

 

Asset-based approaches

 

The value of asset-based analysis a business is equal to the sum of its parts. That is the theory underlying the asset-based approaches to business valuation. The asset approach to business valuation is based on the principle of substitution: no rational investor will pay more for the business assets than the cost of procuring assets of similar economic utility. In contrast to the income-based approaches, which require the valuation professional to make subjective judgments about capitalization or discount rates, the adjusted net book value method is relatively objective. Pursuant to accounting convention, most assets are reported on the books of the subject company at their acquisition value, net of depreciation where applicable. These values must be adjusted to fair market value wherever possible. The value of a company’s intangible assets, such as goodwill, is generally impossible to determine apart from the company’s overall enterprise value. For this reason, the asset-based approach is not the most probative method of determining the value of going business concerns. In these cases, the asset-based approach yields a result that is probably lesser than the fair market value of the business. In considering an asset-based approach, the valuation professional must consider whether the shareholder whose interest is being valued would have any authority to access the value of the assets directly. Shareholders own shares in a corporation, but not its assets, which are owned by the corporation. A controlling shareholder may have the authority to direct the corporation to sell all or part of the assets it owns and to distribute the proceeds to the shareholder(s). The non-controlling shareholder, however, lacks this authority and cannot access the value of the assets. As a result, the value of a corporation’s assets is rarely the most relevant indicator of value to a shareholder who cannot avail himself of that value. Adjusted net book value may be the most relevant standard of value where liquidation is imminent or ongoing; where a company earnings or cash flow are nominal, negative or worth less than its assets; or where net book value is standard in the industry in which the company operates. None of these situations applies to the Company which is the subject of this valuation report. However, the adjusted net book value may be used as a “sanity check” when compared to other methods of valuation, such as the income and market approaches.

 

Market approaches

 

The market approach to business valuation is rooted in the economic principle of competition: that in a free market the supply and demand forces will drive the price of business assets to a certain equilibrium. Buyers would not pay more for the business, and the sellers will not accept less, than the price of a comparable business enterprise. It is similar in many respects to the “comparable sales” method that is commonly used in real estate appraisal. The market price of the stocks of publicly traded companies engaged in the same or a similar line of business, whose shares are actively traded in a free and open market, can be a valid indicator of value when the transactions in which stocks are traded are sufficiently similar to permit meaningful comparison.

 

The difficulty lies in identifying public companies that are sufficiently comparable to the subject company for this purpose. Also, as for a private company, the equity is less liquid (in other words its stocks are less easy to buy or sell) than for a public company, its value is considered to be slightly lower than such a market-based valuation would give

 

Guideline Public Company method

 

The Guideline Public Company method entails a comparison of the subject company to publicly traded companies. The comparison is generally based on published data regarding the public companies’ stock price and earnings, sales, or revenues, which is expressed as a fraction known as a “multiple.” If the guideline public companies are sufficiently similar to each other and the subject company to permit a meaningful comparison, then their multiples should be nearly equal. The public companies identified for comparison purposes should be similar to the subject company in terms of industry, product lines, market, growth, and risk.

 

Transaction Method or Direct Market Data Method

 

Using this method, the valuation analyst may determine market multiples by reviewing published data regarding actual transactions involving either minority or controlling interests in either publicly traded or closely held companies. In judging whether a reasonable basis for comparison exists, the valuation analysis must consider: (1) the similarity of qualitative and quantitative investment and investor characteristics; (2) the extent to which reliable data is known about the transactions in which interests in the guideline companies were bought and sold; and (3) whether or not the price paid for the guideline companies was in an arms-length transaction, or a forced or distressed sale.

 

The most widely used transactional databases include:

 

Institute of Business Appraisers (smaller companies)

BIZCOMPS® (smaller companies)

Pratt’s Stats® (smaller to mid-sized companies)

Public Stats™ (larger companies)

DoneDeals® (larger companies)

Alacra (larger companies)

 

Discounts and premiums

 

The valuation approaches yield the fair market value of the Company as a whole. In valuing a minority, non-controlling interest in a business, however, the valuation professional must consider the applicability of discounts that affect such interests. Discussions of discounts and premiums frequently begin with a review of the “levels of value.” There are three common levels of value: controlling interest, marketable minority, and non-marketable minority. The intermediate level, marketable minority interest, is lesser than the controlling interest level and higher than the non-marketable minority interest level. The marketable minority interest level represents the perceived value of equity interests that are freely traded without any restrictions. These interests are generally traded on the New York Stock Exchange, AMEX, NASDAQ, and other exchanges where there is a ready market for equity securities. These values represent a minority interest in the subject companies – small blocks of stock that represent less than 50% of the company’s equity, and usually much less than 50%. Controlling interest level is the value that an investor would be willing to pay to acquire more than 50% of a company’s stock, thereby gaining the attendant prerogatives of control. Some of the prerogatives of control include: electing directors, hiring and firing the company’s management and determining their compensation; declaring dividends and distributions, determining the company’s strategy and line of business, and acquiring, selling or liquidating the business. This level of value generally contains a control premium over the intermediate level of value, which typically ranges from 25% to 50%. An additional premium may be paid by strategic investors who are motivated by synergistic motives. Non-marketable, minority level is the lowest level on the chart, representing the level at which non-controlling equity interests in private companies are generally valued or traded. This level of value is discounted because no ready market exists in which to purchase or sell interests. Private companies are less “liquid” than publicly-traded companies, and transactions in private companies take longer and are more uncertain. Between the intermediate and lowest levels of the chart, there are restricted shares of publicly-traded companies. Despite a growing inclination of the IRS and Tax Courts to challenge valuation discounts , Shannon Pratt suggested in a scholarly presentation recently that valuation discounts are actually increasing as the differences between public and private companies is widening . Publicly-traded stocks have grown more liquid in the past decade due to rapid electronic trading, reduced commissions, and governmental deregulation. These developments have not improved the liquidity of interests in private companies, however. Valuation discounts are multiplicative, so they must be considered in order. Control premiums and their inverse, minority interest discounts, are considered before marketability discounts are applied.

 

Discount for lack of control

 

The first discount that must be considered is the discount for lack of control, which in this instance is also a minority interest discount. Minority interest discounts are the inverse of control premiums, to which the following mathematical relationship exists: MID = 1 – [ 1 / (1 + CP)] The most common source of data regarding control premiums is the Control Premium Study, published annually by Mergerstat since 1972. Mergerstat compiles data regarding publicly announced mergers, acquisitions and divestitures involving 10% or more of the equity interests in public companies, where the purchase price is $1 million or more and at least one of the parties to the transaction is a U.S. entity. Mergerstat defines the “control premium” as the percentage difference between the acquisition price and the share price of the freely-traded public shares five days prior to the announcement of the M&A transaction. While it is not without valid criticism, Mergerstat control premium data (and the minority interest discount derived therefrom) is widely accepted within the valuation profession.

 

Discount for lack of marketability

 

Another factor to be considered in valuing closely held companies is the marketability of an interest in such businesses. Marketability is defined as the ability to convert the business interest into cash quickly, with minimum transaction and administrative costs, and with a high degree of certainty as to the amount of net proceeds. There is usually a cost and a time lag associated with locating interested and capable buyers of interests in privately-held companies, because there is no established market of readily-available buyers and sellers. All other factors being equal, an interest in a publicly traded company is worth more because it is readily marketable. Conversely, an interest in a private-held company is worth less because no established market exists. The IRS Valuation Guide for Income, Estate and Gift Taxes, Valuation Training for Appeals Officers acknowledges the relationship between value and marketability, stating: “Investors prefer an asset which is easy to sell, that is, liquid.” The discount for lack of control is separate and distinguishable from the discount for lack of marketability. It is the valuation professional’s task to quantify the lack of marketability of an interest in a privately-held company. Because, in this case, the subject interest is not a controlling interest in the Company, and the owner of that interest cannot compel liquidation to convert the subject interest to cash quickly, and no established market exists on which that interest could be sold, the discount for lack of marketability is appropriate. Several empirical studies have been published that attempt to quantify the discount for lack of marketability. These studies include the restricted stock studies and the pre-IPO studies. The aggregate of these studies indicate average discounts of 35% and 50%, respectively. Some experts believe the Lack of Control and Marketabilty discounts can aggregate discounts for as much as ninety percent of a Company’s fair market value, specifically with family owned companies.

 

Restricted stock studies

 

Restricted stocks are equity securities of public companies that are similar in all respects to the freely traded stocks of those companies except that they carry a restriction that prevents them from being traded on the open market for a certain period of time, which is usually one year (two years prior to 1990). This restriction from active trading, which amounts to a lack of marketability, is the only distinction between the restricted stock and its freely-traded counterpart. Restricted stock can be traded in private transactions and usually do so at a discount. The restricted stock studies attempt to verify the difference in price at which the restricted shares trade versus the price at which the same unrestricted securities trade in the open market as of the same date. The underlying data by which these studies arrived at their conclusions has not been made public. Consequently, it is not possible when valuing a particular company to compare the characteristics of that company to the study data. Still, the existence of a marketability discount has been recognized by valuation professionals and the Courts, and the restricted stock studies are frequently cited as empirical evidence. Notably, the lowest average discount reported by these studies was 26% and the highest average discount was 45%.

 

Option pricing

 

In addition to the restricted stock studies, U.S. publicly traded companies are able to sell stock to offshore investors (SEC Regulation S, enacted in 1990) without registering the shares with the Securities and Exchange Commission. The offshore buyers may resell these shares in the United States, still without having to register the shares, after holding them for just 40 days. Typically, these shares are sold for 20% to 30% below the publicly traded share price. Some of these transactions have been reported with discounts of more than 30%, resulting from the lack of marketability. These discounts are similar to the marketability discounts inferred from the restricted and pre-IPO studies, despite the holding period being just 40 days. Studies based on the prices paid for options have also confirmed similar discounts. If one holds restricted stock and purchases an option to sell that stock at the market price (a put), the holder has, in effect, purchased marketability for the shares. The price of the put is equal to the marketability discount. The range of marketability discounts derived by this study was 32% to 49%.

 

Pre-IPO studies

 

Another approach to measure the marketability discount is to compare the prices of stock offered in initial public offerings (IPOs) to transactions in the same company’s stocks prior to the IPO. Companies that are going public are required to disclose all transactions in their stocks for a period of three years prior to the IPO. The pre-IPO studies are the leading alternative to the restricted stock stocks in quantifying the marketability discount. The pre-IPO studies are sometimes criticized because the sample size is relatively small, the pre-IPO transactions may not be arm’s length, and the financial structure and product lines of the studied companies may have changed during the three year pre-IPO window.

 

Applying the studies

 

The studies confirm what the marketplace knows intuitively: Investors covet liquidity and loathe obstacles that impair liquidity. Prudent investors buy illiquid investments only when there is a sufficient discount in the price to increase the rate of return to a level which brings risk-reward back into balance. The referenced studies establish a reasonable range of valuation discounts from the mid-30%s to the low 50%s. The more recent studies appeared to yield a more conservative range of discounts than older studies, which may have suffered from smaller sample sizes. Another method of quantifying the lack of marketability discount is the Quantifying Marketability Discounts Model (QMDM).

 

DISCOUNTED CASH FLOW

 

In finance, the discounted cash flow (or DCF) approach describes a method to value a project, company, or financial asset using the concepts of the time value of money. All future cash flows are estimated and discounted to give them a present value. The discount rate used is generally the appropriate cost of capital, and incorporates judgments of the uncertainty (riskiness) of the future cash flows.

 

FV=PV (1+i)n

 

DPV=FV/(1+i)n

 

COST OF CAPITAL

 

The cost of capital for a firm is a weighted sum of the cost of equity and the cost of debt (see Capital investment decisions). It is also known as the “Hurdle Rate” or “Discount Rate”.

 

Capital (money) used to fund a business should earn returns for the capital owner who risked his/her saved money. For an investment to be worthwhile the projected return on capital must be greater than the cost of capital. Otherwise stated, the risk-adjusted return on capital (that is, incorporating not just the projected returns, but the probabilities of those projections) must be higher than the cost of capital.

 

The cost of debt is relatively simple to calculate, as it is composed of the rate of interest paid. In practice, the interest-rate paid by the company will include the risk-free rate plus a risk component, which itself incorporates a probable rate of default (and amount of recovery given default). For companies with similar risk or credit ratings, the interest rate is largely exogenous.

 

Cost of equity is more challenging to calculate as equity does not pay a set return to its investors. Similar to the cost of debt, the cost of equity is broadly defined as the risk-weighted projected return required by investors, where the return is largely unknown. The cost of equity is therefore inferred by comparing the investment to other investments with similar risk profiles to determine the “market” cost of equity.

 

The cost of capital is often used as the discount rate, the rate at which projected cash flow will be discounted to give a present value or net present value.

 

Cost of debt

 

The cost of debt is computed by taking the rate on a non-defaulting bond whose duration matches the term structure of the corporate debt, then adding a default premium. This default premium will rise as the amount of debt increases (since the risk rises as the amount of debt rises). Since in most cases debt expense is a deductible expense, the cost of debt is computed as an after tax cost to make it comparable with the cost of equity (earnings are after-tax as well). Thus, for profitable firms, debt is discounted by the tax rate. Basically this is used for large corporations only.

 

Cost of equity

 

Cost of equity = Risk free rate of return + Premium expected for risk

 

Expected return

 

The expected return can be calculated as the “dividend capitalization model”, which is (dividend per share / price per share) + growth rate of dividends (that is, dividend yield + growth rate of dividends).

 

Capital asset pricing model

 

The capital asset pricing model (CAPM) is used in finance to determine a theoretically appropriate price of an asset such as a security. The expected return on equity according to the capital asset pricing model. The market risk is normally characterized by the ? parameter. Thus, the investors would expect (or demand) to receive:

 

WEIGHTED AVERAGE COST OF CAPITAL

 

The Weighted Average Cost of Capital (WACC) is used in finance to measure a firm’s cost of capital.

 

The total capital for a firm is the value of its equity (for a firm without outstanding warrants and options, this is the same as the company’s market capitalization) plus the cost of its debt (the cost of debt should be continually updated as the cost of debt changes as a result of interest rate changes). Notice that the “equity” in the debt to equity ratio is the market value of all equity, not the shareholders’ equity on the balance sheet.

 

Calculation of WACC is an iterative procedure which requires estimation of the fair market value of equity capital

 

CAPITAL STRUCTURE

 

Because of tax advantages on debt issuance, it will be cheaper to issue debt rather than new equity (this is only true for profitable firms, tax breaks are available only to profitable firms). At some point, however, the cost of issuing new debt will be greater than the cost of issuing new equity. This is because adding debt increases the default risk - and thus the interest rate that the company must pay in order to borrow money. By utilizing too much debt in its capital structure, this increased default risk can also drive up the costs for other sources (such as retained earnings and preferred stock) as well. Management must identify the “optimal mix” of financing – the capital structure where the cost of capital is minimized so that the firms value can be maximized.

 

MODIGLIANI-MILLER THEOREM

 

If there were no tax advantages for issuing debt, and equity could be freely issued, Miller and Modigliani showed that the value of a leveraged firm and the value of an unleveraged firm should be the same.

 

INTEREST

 

Interest is a fee paid on borrowed capital. Assets lent include money, shares, consumer goods through hire purchase, major assets such as aircraft, and even entire factories in finance lease arrangements. The interest is calculated upon the value of the assets in the same manner as upon money. Interest can be thought of as “rent on money”.

 

The fee is compensation to the lender for foregoing other useful investments that could have been made with the loaned money. Instead of the lender using the assets directly, they are advanced to the borrower. The borrower then enjoys the benefit of using the assets ahead of the effort required to obtain them, while the lender enjoys the benefit of the fee paid by the borrower for the privilege. The amount lent, or the value of the assets lent, is called the principal. This principal value is held by the borrower on credit. Interest is therefore the price of credit, not the price of money as it is commonly - and mistakenly - believed to be. The percentage of the principal that is paid as a fee (the interest), over a certain period of time, is called the interest rate.

 

Interest rates and credit risk

 

It is increasingly recognized that the business cycle, interest rates and credit risk are tightly interrelated. The Jarrow-Turnbull model was the first model of credit risk which explicitly had random interest rates at its core. Lando (2004), Darrell Duffie and Singleton (2003), and van Deventer and Imai (2003) discuss interest rates when the issuer of the interest-bearing instrument can default.

 

Money and inflation

Loans, bonds, and shares have some of the characteristics of money and are included in the broad money supply.

 

By setting i*n, the government institution can affect the markets to alter the total of loans, bonds and shares issued. Generally speaking, a higher real interest rate reduces the broad money supply.

 

Open market operations in the United States

 

The Federal Reserve (often referred to as ‘The Fed’) implements monetary policy largely by targeting the federal funds rate. This is the rate that banks charge each other for overnight loans of federal funds. Federal funds are the reserves held by banks at the Fed.

 

Open market operations are one tool within monetary policy implemented by the Federal Reserve to steer short-term interest rates. Using the power to buy and sell treasury securities, the Open Market Desk at the Federal Reserve Bank of New York can supply the market with dollars by purchasing T-notes, hence increasing the nation’s money supply. By increasing the money supply or Aggregate Supply of Funding (ASF), interest rates will fall due to the excess of dollars banks will end up with in their reserves. Excess reserves may be lent in the Fed funds market to other banks, thus driving down rates.

 

Credit spread options:  credit call spread is a “bearish” call spread, which has more premium on the short call.  A credit put spread is a “bullish” put spread and has more premium on the short put.

 

Credit spread (bond): In finance, a credit spread is the difference in yield between different securities due to different credit quality. The credit spread reflects the additional net yield an investor can earn from a security with more credit risk relative to one with less credit risk. The credit spread of a particular security is often quoted in relation to the yield on a credit risk-free benchmark security or reference rate.

 

RISK MODELING

 

Risk modeling refers to the use of formal econometric techniques to determine the aggregate risk in a financial portfolio. Risk modeling is one of many subtasks within the broader area of financial modeling.

 

Risk modeling uses a variety of techniques including market risk, Value-at-Risk (VaR), Historical Simulation (HS), or Extreme Value Theory (EVT) in order to analyze a portfolio and make forecasts of the likely losses that would be incurred for a variety of risks. Such risks are typically grouped into credit risk, liquidity risk, interest rate risk, and operational risk categories.

 

Many large financial intermediary firms use risk modeling to help portfolio managers assess the amount of capital reserves to maintain, and to help guide their purchases and sales of various classes of financial assets.

 

Formal risk modeling is required under the Basel II proposal for all the major international banking institutions by the various national depository institution regulators.

 

Quantitative risk analysis and modeling have become important in the light of corporate scandals in the past few years (most notably, Enron), Basel II, the revised FAS 123R and the Sarbanes-Oxley Act. In the past, risk analysis was done qualitatively but now with the advent of powerful computing software, quantitative risk analysis can be done quickly and effortlessly.

 

PORTFOLIO PROBLEMS

 

 

In finance, a portfolio is an appropriate mix of or collection of investments held by an institution or a private individual. In building up an investment portfolio a financial institution will typically conduct its own investment analysis, whilst a private individual may make use of the services of a financial advisor or a financial institution which offers portfolio management services. Holding a portfolio is part of an investment and risk-limiting strategy called diversification. By owning several assets, certain types of risk (in particular specific risk) can be reduced. The assets in the portfolio could include stocks, bonds, options, warrants, gold certificates, real estate, futures contracts, production facilities, or any other item that is expected to retain its value.

 

Portfolio management involves deciding what assets to include in the portfolio, given the goals of the portfolio owner and changing economic conditions. Selection involves deciding what assets to purchase, how many to purchase, when to purchase them, and what assets to divest. These decisions always involve some sort of performance measurement, most typically expected return on the portfolio, and the risk associated with this return (i.e. the standard deviation of the return). Typically the expected return from portfolios of different asset bundles are compared.

 

Porfolio formation

Many strategies have been developed to form a portfolio.

 

Ø       equally-weighted portfolio

Ø       capitalization-weighted portfolio

Ø       price-weighted portfolio

Ø       optimal portfolio (for which the Sharpe ratio is highest)

 

VALUATION OF OPTIONS

 

Black–Scholes:

 

The term Black–Scholes refers to three closely related concepts:

 

Ø       The Black–Scholes model is a mathematical model of the market for an equity, in which the equity’s price is a stochastic process.

Ø       The Black–Scholes PDE is a partial differential equation which (in the model) must be satisfied by the price of a derivative on the equity.

Ø       The Black–Scholes formula is the result obtained by solving the Black-Scholes PDE for European put and call options.

 

Binomial options pricing model: In finance, the binomial options pricing model (BOPM) provides a generalisable numerical method for the valuation of options. The binomial model was first proposed by Cox, Ross and Rubinstein (1979). Essentially, the model uses a “discrete-time” model of the varying price over time of the underlying financial instrument. Option valuation is then computed via application of the risk neutrality assumption over the life of the option, as the price of the underlying instrument evolves.

 

 

Monte Carlo option model: In mathematical finance, a Monte Carlo option model uses Monte Carlo methods to calculate the value of an option with multiple sources of uncertainty or with complicated features.

 

REAL OPTIONS ANALYSIS

 

In corporate finance, real options analysis or ROA applies put option and call option valuation techniques to capital budgeting decisions.[1]

 

A real option is the right, but not the obligation, to undertake some business decision, typically the option to make a capital investment. For example, the opportunity to invest in the expansion of a firm’s factory is a real option. In contrast to financial options, a real option is not often tradeable—e.g. the factory owner cannot sell the right to extend his factory to another party, only he can make this decision; however, some real options can be sold, e.g., ownership of a vacant lot of land is a real option to develop that land in the future. Some real options are proprietory (owned or exercisable by a single individual or a company); others are shared (can be exercised by many parties). Therefore, a project may have a portfolio of embedded real options; some of them can be mutually exclusive.

 

The terminology “real option” is relatively new, whereas business operators have been making capital investment decisions for centuries. However, the description of such opportunities as real options has occurred at the same time as thinking about such decisions in new, more analytically-based, ways. As such, the terminology “real option” is closely tied to these new methods. The term “real option” was coined by Professor Stewart Myers at the MIT Sloan School of Management; this happened most likely around 1977.

 

The concept of real options was popularized by Michael J. Mauboussin, the chief U.S. investment strategist for Credit Suisse First Boston and an adjunct professor of finance at the Columbia School of Business. Mauboussin uses real options in part to explain the gap between how the stock market prices some businesses and the “intrinsic value” for those businesses as calculated by traditional financial analysis, specifically discounted cash flows.

 

Additionally, with real option analysis, uncertainty inherent in investment projects is usually accounted for by risk-adjusting probabilities (a technique known as the equivalent martingale approach). Cash flows can then be discounted at the risk-free rate. With regular DCF analysis, on the other hand, this uncertainty is accounted for by adjusting the discount rate, using e.g. the cost of capital) or the cash flows (using certainty equivalents). These methods normally do not properly account for changes in risk over a project’s lifecycle and fail to appropriately adapt the risk adjustment. More importantly, the real options approach forces decision makers to be more explicit about the assumptions underlying their projections.

 

Generally, the most widely used methods are: Closed form solutions, partial differential equations, and the binomial lattices. In business strategy, real options have been advanced by the construction of option space, where volatility is compared with value-to-cost, NPVq. Latest advances in real option valuation are models that incorporate fuzzy logic and option valuation in fuzzy real option valuation models.

 

Real options are a field of academic research, and at the present one of the leading names in academic real options is Professor Lenos Trigeorgis (University of Cyprus). An academic conference on real options is organized yearly (Annual International Conference on Real Options).



By: Kiran Kumar Cherupalli

About the Author:



Marlene

neural network

Tuesday, November 20th, 2007
 

What is a Neural Network?

 

First of all, when we are talking about a neural network, we should more properly say "artificial neural network" (ANN), because that is what we mean most of the time. Biological neural networks are much more complicated than the mathematical models we use for ANNs. But it is customary to be lazy and drop the "A" or the "artificial".

An Artificial Neural Network (ANN) is an information processing paradigm that is inspired by the way biological nervous systems, such as the brain, process information. The key element of this paradigm is the novel structure of the information processing system. It is composed of a large number of highly interconnected processing elements (neurons) working in unison to solve specific problems. ANNs, like people, learn by example. An ANN is configured for a specific application, such as pattern recognition or data classification, through a learning process. Learning in biological systems involves adjustments to the synaptic connections that exist between the neurons. This is true of ANNs as well.

 

 

Historical Background of Neural Networks

 

Neural network simulations appear to be a recent development. However, this field was established before the advent of computers, and has survived at least one major setback and several eras.

Many importand advances have been boosted by the use of inexpensive computer emulations. Following an initial period of enthusiasm, the field survived a period of frustration and disrepute. During this period when funding and professional support was minimal, important advances were made by relatively few reserchers. These pioneers were able to develop convincing technology which surpassed the limitations identified by Minsky and Papert. Minsky and Papert, published a book (in 1969) in which they summed up a general feeling of frustration (against neural networks) among researchers, and was thus accepted by most without further analysis. Currently, the neural network field enjoys a resurgence of interest and a corresponding increase in funding.

 

The history of neural networks that was described above can be divided into several periods:

 

First Attempts: There were some initial simulations using formal logic. McCulloch and Pitts (1943) developed models of neural networks based on their understanding of neurology. These models made several assumptions about how neurons worked. Their networks were based on simple neurons which were considered to be binary devices with fixed thresholds. The results of their model were simple logic functions such as "a or b" and "a and b". Another attempt was by using computer simulations. Two groups (Farley and Clark, 1954; Rochester, Holland, Haibit and Duda, 1956). The first group (IBM reserchers) maintained closed contact with neuroscientists at McGill University. So whenever their models did not work, they consulted the neuroscientists. This interaction established a multidiscilinary trend which continues to the present day.

 

Promising & Emerging Technology: Not only was neroscience influential in the development of neural networks, but psychologists and engineers also contributed to the progress of neural network simulations. Rosenblatt (1958) stirred considerable interest and activity in the field when he designed and developed the Perceptron. The Perceptron had three layers with the middle layer known as the association layer.This system could learn to connect or associate a given input to a random output unit.

Another system was the ADALINE (ADAptive LInear Element) which was developed in 1960 by Widrow and Hoff (of Stanford University). The ADALINE was an analogue electronic device made from simple components. The method used for learning was different to that of the Perceptron, it employed the Least-Mean-Squares (LMS) learning rule.

 

Period of Frustration & Disrepute: In 1969 Minsky and Papert wrote a book in which they generalised the limitations of single layer Perceptrons to multilayered systems. In the book they said: "…our intuitive judgment that the extension (to multilayer systems) is sterile". The significant result of their book was to eliminate funding for research with neural network simulations. The conclusions supported the disenhantment of reserchers in the field. As a result, considerable prejudice against this field was activated.

 

Innovation: Although public interest and available funding were minimal, several researchers continued working to develop neuromorphically based computaional methods for problems such as pattern recognition.

During this period several paradigms were generated which modern work continues to enhance.Grossberg’s (Steve Grossberg and Gail Carpenter in 1988) influence founded a school of thought which explores resonating algorithms. They developed the ART (Adaptive Resonance Theory) networks based on biologically plausible models. Anderson and Kohonen developed associative techniques independent of each other. Klopf (A. Henry Klopf) in 1972, developed a basis for learning in artificial neurons based on a biological principle for neuronal learning called heterostasis.

Werbos (Paul Werbos 1974) developed and used the back-propagation learning method, however several years passed before this approach was popularized. Back-propagation nets are probably the most well known and widely applied of the neural networks today. In essence, the back-propagation net. is a Perceptron with multiple layers, a different thershold function in the artificial neuron, and a more robust and capable learning rule.

Amari (A. Shun-Ichi 1967) was involved with theoretical developments: he published a paper which established a mathematical theory for a learning basis (error-correction method) dealing with adaptive patern classification. While Fukushima (F. Kunihiko) developed a step wise trained multilayered neural network for interpretation of handwritten characters. The original network was published in 1975 and was called the Cognitron.

 

Re-Emergence: Progress during the late 1970s and early 1980s was important to the re-emergence on interest in the neural network field. Several factors influenced this movement. For example, comprehensive books and conferences provided a forum for people in diverse fields with specialized technical languages, and the response to conferences and publications was quite positive. The news media picked up on the increased activity and tutorials helped disseminate the technology. Academic programs appeared and courses were inroduced at most major Universities (in US and Europe). Attention is now focused on funding levels throughout Europe, Japan and the US and as this funding becomes available, several new commercial with applications in industry and finacial institutions are emerging.

Today: Significant progress has been made in the field of neural networks-enough to attract a great deal of attention and fund further research. Advancement beyond current commercial applications appears to be possible, and research is advancing the field on many fronts. Neurally based chips are emerging and applications to complex problems developing. Clearly, today is a period of transition for neural network technology.

 

Why use neural networks?

 

Neural networks, with their remarkable ability to derive meaning from complicated or imprecise data, can be used to extract patterns and detect trends that are too complex to be noticed by either humans or other computer techniques. A trained neural network can be thought of as an "expert" in the category of information it has been given to analyse. This expert can then be used to provide projections given new situations of interest and answer "what if" questions.

Other advantages include:

 

Adaptive learning: An ability to learn how to do tasks based on the data given for training or initial experience. Self-Organisation: An ANN can create its own organisation or representation of the information it receives during learning time. Real Time Operation: ANN computations may be carried out in parallel, and special hardware devices are being designed and manufactured which take advantage of this capability. Fault Tolerance via Redundant Information Coding: Partial destruction of a network leads to the corresponding degradation of performance. However, some network capabilities may be retained even with major network damage.

 

          Neural networks versus conventional computers

 

Neural networks take a different approach to problem solving than that of conventional computers. Conventional computers use an algorithmic approach i.e. the computer follows a set of instructions in order to solve a problem. Unless the specific steps that the computer needs to follow are known the computer cannot solve the problem. That restricts the problem solving capability of conventional computers to problems that we already understand and know how to solve. But computers would be so much more useful if they could do things that we don’t exactly know how to do.

Neural networks process information in a similar way the human brain does. The network is composed of a large number of highly interconnected processing elements(neurones) working in parallel to solve a specific problem. Neural networks learn by example. They cannot be programmed to perform a specific task. The examples must be selected carefully otherwise useful time is wasted or even worse the network might be functioning incorrectly. The disadvantage is that because the network finds out how to solve the problem by itself, its operation can be unpredictable.

On the other hand, conventional computers use a cognitive approach to problem solving; the way the problem is to solved must be known and stated in small unambiguous instructions. These instructions are then converted to a high level language program and then into machine code that the computer can understand. These machines are totally predictable; if anything goes wrong is due to a software or hardware fault.

Neural networks and conventional algorithmic computers are not in competition but complement each other. There are tasks are more suited to an algorithmic approach like arithmetic operations and tasks that are more suited to neural networks. Even more, a large number of tasks, require systems that use a combination of the two approaches (normally a conventional computer is used to supervise the neural network) in order to perform at maximum efficiency.

 

 

 

 

 

 

Neural Networks in Practice

 

Given this description of neural networks and how they work, what real world applications are they suited for? Neural networks have broad applicability to real world business problems. In fact, they have already been successfully applied in many industries.

 

Since neural networks are best at identifying patterns or trends in data, they are well suited for prediction or forecasting needs including:

 

 

sales forecasting

industrial process control

customer research

data validation

risk management

target marketing

But to give you some more specific examples; ANN are also used in the following specific paradigms: recognition of speakers in communications; diagnosis of hepatitis; recovery of telecommunications from faulty software; interpretation of multimeaning Chinese words; undersea mine detection; texture analysis; three-dimensional object recognition; handwritten word recognition; and facial recognition.

 

 

 

 

 

Human and Artificial Neurones - investigating the similarities

 

How the Human Brain Learns?

 

Much is still unknown about how the brain trains itself to process information, so theories abound. In the human brain, a typical neuron collects signals from others through a host of fine structures called dendrites. The neuron sends out spikes of electrical activity through a long, thin stand known as an axon, which splits into thousands of branches. At the end of each branch, a structure called a synapse converts the activity from the axon into electrical effects that inhibit or excite activity from the axon into electrical effects that inhibit or excite activity in the connected neurones. When a neuron receives excitatory input that is sufficiently large compared with its inhibitory input, it sends a spike of electrical activity down its axon. Learning occurs by changing the effectiveness of the synapses so that the influence of one neuron on another changes.

 

 

 

 

 

 

Components of a neuron

 

 

 

 

 

The synapse

 

  From Human Neurones to Artificial Neurones

 

We conduct these neural networks by first trying to deduce the essential features of neurones and their interconnections. We then typically program a computer to simulate these features. However because our knowledge of neurones is incomplete and our computing power is limited, our models are necessarily gross idealisations of real networks of neurones.

 

The neuron model

 

Architecture of neural networks

Feed-forward networks

 Feed-forward ANNs  allow signals to travel one way only; from input to output. There is no feedback (loops) i.e. the output of any layer does not affect that same layer. Feed-forward ANNs tend to be straight forward networks that associate inputs with outputs. They are extensively used in pattern recognition. This type of organisation is also referred to as bottom-up or top-down.

 

Feedback networks

Feedback networks (figure 1) can have signals travelling in both directions by introducing loops in the network. Feedback networks are very powerful and can get extremely complicated. Feedback networks are dynamic; their ’state’ is changing continuously until they reach an equilibrium point. They remain at the equilibrium point until the input changes and a new equilibrium needs to be found. Feedback architectures are also referred to as interactive or recurrent, although the latter term is often used to denote feedback connections in single-layer organisations.



 

 

 

 

Applications of neural networks

Neural Networks in Practice

Given this description of neural networks and how they work, what real world applications are they suited for? Neural networks have broad applicability to real world business problems. In fact, they have already been successfully applied in many industries.

Since neural networks are best at identifying patterns or trends in data, they are well suited for prediction or forecasting needs including:

sales forecasting

industrial process control

customer research

data validation

risk management

target marketing

But to give you some more specific examples; ANN are also used in the following specific paradigms: recognition of speakers in communications; diagnosis of hepatitis; recovery of telecommunications from faulty software; interpretation of multimeaning Chinese words; undersea mine detection; texture analysis; three-dimensional object recognition; hand-written word recognition; and facial recognition.

Neural networks in medicine

 

Artificial Neural Networks (ANN) are currently a ‘hot’ research area in medicine and it is believed that they will receive extensive application to biomedical systems in the next few years. At the moment, the research is mostly on modelling parts of the human body and recognising diseases from various scans (e.g. cardiograms, CAT scans, ultrasonic scans, etc.).

Neural networks are ideal in recognising diseases using scans since there is no need to provide a specific algorithm on how to identify the disease. Neural networks learn by example so the details of how to recognise the disease are not needed. What is needed is a set of examples that are representative of all the variations of the disease. The quantity of examples is not as important as the ‘quantity’. The examples need to be selected very carefully if the system is to perform reliably and efficiently.

Modelling and Diagnosing the Cardiovascular System

Neural Networks are used experimentally to model the human cardiovascular system. Diagnosis can be achieved by building a model of the cardiovascular system of an individual and comparing it with the real time physiological measurements taken from the patient. If this routine is carried out regularly, potential harmful medical conditions can be detected at an early stage and thus make the process of combating the disease much easier.

A model of an individual’s cardiovascular system must mimic the relationship among physiological variables (i.e., heart rate, systolic and diastolic blood pressures, and breathing rate) at different physical activity levels. If a model is adapted to an individual, then it becomes a model of the physical condition of that individual. The simulator will have to be able to adapt to the features of any individual without the supervision of an expert. This calls for a neural network.

Another reason that justifies the use of ANN technology, is the ability of ANNs to provide sensor fusion which is the combining of values from several different sensors. Sensor fusion enables the ANNs to learn complex relationships among the individual sensor values, which would otherwise be lost if the values were individually analysed. In medical modelling and diagnosis, this implies that even though each sensor in a set may be sensitive only to a specific physiological variable, ANNs are capable of detecting complex medical conditions by fusing the data from the individual biomedical sensors.

Electronic noses

ANNs are used experimentally to implement electronic noses. Electronic noses have several potential applications in telemedicine. Telemedicine is the practice of medicine over long distances via a communication link. The electronic nose would identify odours in the remote surgical environment. These identified odours would then be electronically transmitted to another site where an door generation system would recreate them. Because the sense of smell can be an important sense to the surgeon, telesmell would enhance telepresent surgery.

For more information on telemedicine and telepresent surgery

 

Electronic noses

ANNs are used experimentally to implement electronic noses. Electronic noses have several potential applications in telemedicine. Telemedicine is the practice of medicine over long distances via a communication link. The electronic nose would identify odours in the remote surgical environment. These identified odours would then be electronically transmitted to another site where an door generation system would recreate them. Because the sense of smell can be an important sense to the surgeon, telesmell would enhance telepresent surgery.

For more information on telemedicine and telepresent surgery

    Neural Networks in business

 

 

Business is a diverted field with several general areas of specialisation such as accounting or financial analysis. Almost any neural network application would fit into one business area or financial analysis.

There is some potential for using neural networks for business purposes, including resource allocation and scheduling. There is also a strong potential for using neural networks for database mining, that is, searching for patterns implicit within the explicitly stored information in databases. Most of the funded work in this area is classified as proprietary. Thus, it is not possible to report on the full extent of the work going on. Most work is applying neural networks, such as the Hopfield-Tank network for optimization and scheduling.

    Marketing

There is a marketing application which has been integrated with a neural network system. The Airline Marketing Tactician (a trademark abbreviated as AMT) is a computer system made of various intelligent technologies including expert systems. A feedforward neural network is integrated with the AMT and was trained using back-propagation to assist the marketing control of airline seat allocations. The adaptive neural approach was amenable to rule expression. Additionaly, the application’s environment changed rapidly and constantly, which required a continuously adaptive solution. The system is used to monitor and recommend booking advice for each departure. Such information has a direct impact on the profitability of an airline and can provide a technological advantage for users of the system. [Hutchison & Stephens, 1987]

While it is significant that neural networks have been applied to this problem, it is also important to see that this intelligent technology can be integrated with expert systems and other approaches to make a functional system. Neural networks were used to discover the influence of undefined interactions by the various variables. While these interactions were not defined, they were used by the neural system to develop useful conclusions. It is also noteworthy to see that neural networks can influence the bottom line.

 

 

 

Are there any limits to Neural Networks?

 

The major issues of concern today are the scalability problem, testing, verification, and integration of neural network systems into the modern environment. Neural network programs sometimes become unstable when applied to larger problems. The defence, nuclear and space industries are concerned about the issue of testing and verification. The mathematical theories used to guarantee the performance of an applied neural network are still under development. The solution for the time being may be to train and test these intelligent systems much as we do for humans. Also there are some more practical problems like:

the operational problem encountered when attempting to simulate the parallelism of neural networks. Since the majority of neural networks are simulated on sequential machines, giving rise to a very rapid increase in processing time requirements as size of the problem expands.

Solution: implement neural networks directly in hardware, but these need a lot of development still. instability to explain any results that they obtain. Networks function as "black boxes" whose rules of operation are completely unknown

 

 

The Future

Because gazing into the future is somewhat like gazing into a crystal ball, so it is better to quote some "predictions". Each prediction rests on some sort of evidence or established trend which, with extrapolation, clearly takes us into a new realm.

Prediction 1:

Neural Networks will fascinate user-specific systems for education, information processing, and entertainment. "Alternative ralities", produced by comprehensive environments, are attractive in terms of their potential for systems control, education, and entertainment. This is not just a far-out research trend, but is something which is becoming an increasing part of our daily existence, as witnessed by the growing interest in comprehensive "entertainment centers" in each home.

This "programming" would require feedback from the user in order to be effective but simple and "passive" sensors (e.g fingertip sensors, gloves, or wristbands to sense pulse, blood pressure, skin ionisation, and so on), could provide effective feedback into a neural control system. This could be achieved, for example, with sensors that would detect pulse, blood pressure, skin ionisation, and other variables which the system could learn to correlate with a person’s response state.

Prediction 2:

Neural networks, integrated with other artificial intelligence technologies, methods for direct culture of nervous tissue, and other exotic technologies such as genetic engineering, will allow us to develop radical and exotic life-forms whether man, machine, or hybrid.

Prediction 3:

Neural networks will allow us to explore new realms of human capabillity realms previously available only with extensive training and personal discipline. So a specific state of consiously induced neurophysiologically observable awareness is necessary in order to facilitate a man machine system interface.

 

Conclusion

The computing world has a lot to gain fron neural networks. Their ability to learn by example makes them very flexible and powerful. Furthermore there is no need to devise an algorithm in order to perform a specific task; i.e. there is no need to understand the internal mechanisms of that task. They are also very well suited for real time systems because of their fast responseand computational times which are due to their parallel architecture.

Neural networks also contribute to other areas of research such as neurology and psychology. They are regularly used to model parts of living organisms and to investigate the internal mechanisms of the brain.

Perhaps the most exciting aspect of neural networks is the possibility that some day ‘consious’ networks might be produced. There is a number of scientists arguing that conciousness is a ‘mechanical’ property and that ‘consious’ neural networks are a realistic possibility.

Finally, I would like to state that even though neural networks have a huge potential we will only get the best of them when they are intergrated with computing, AI, fuzzy logic and related subjects.

 



By: Vish Khandelwal

About the Author:

vish khandelwal



Sherry

Current Advancements in Audio Visual Field

Friday, November 16th, 2007


There are times when technology ventures into the realm of the weird. Yet, the spin-offs ultimately benefit mankind in unimaginable ways. Today’s consumer electronic news yields two such instances- both exotic and incredible. One is a humanoid from Computer Graphics International and the other is the robotic snake, the brainchild of Dr Gavin Miller.

The idea of a humanoid probably germinated around 250 BC. Today’s humanoid robot does not need any maintenance, learns on its own and can coexist in a safe human environment. But, that’s not all. The technology to develop these humanoids with fuzzy logic has benefited the huge industry of gaming computers. There is more. The spin-offs are most impressive. Today’s Harry Potter films among others are a result of this technology. Enter Emily O’Brien the actress. The Emily Project resulted in a human like animation which could copy Emily’s live video images using diverse parameters like hand-eye-body coordination, lip synchronization and the like.The technology will also benefit the Security Industry in recognizing faces.

Consumer electronic news talks about the robotic snake. This is an incredible marvel to help locate survivors post earthquake, house collapse or any such mishaps. The technology actually replicates not only the slithering motion of the snake but also its sharp sensory perception for changes in external conditions. The mechanical snake has its own computers and power source. Normally, it is powered and guided through an umbilical cord which can be disengaged and the mode changed to radio control when required. The need for artificial intelligence having been obviated, the snake has a sensitive camera and sound transducers. It also has a ranging facility involving IR for sensing body heat. The scales on its body permit an all terrain travel. The prime ability is perhaps the empowerment of explosive technology in order that the robot can make a 10 cm hole after retreating and enter the hole after clearing up the debris!!

Today each technological find will be assessed on its utility scale. The robotic snake and the humanoid have justified their technologies on account of the widening of horizon which has resulted.



By: Anirban Bhattacharya

About the Author:

I webmaster of http://www.digitaltrends.com provides consumer electronics reviews, Technology news, technology reviews, electronics guides, Technology Videos, Electronics Guides including videos, podcasts, newsletters, discussion forums, unbiased reviews, and up-to-the-minute information on everything that is latest.



Randy

natural language processing outsourcing?

Friday, November 16th, 2007
david w asked:



For staffing.


Kelly

Where can I find these Computer Science classes online?

Monday, November 5th, 2007
Jimstein asked:



An online msee program but not online data structuresalgorithms ii this course presents algorithm design techniques such as dynamic.

For later courses are available but not online data structuresalgorithms ii this course presents algorithm design techniques for later courses are available but not online data structures such as logic theoremproving language theory graph algorithms.


Ann