Financial economics/Tutorials

Financial market hypotheses

Efficient market hypothesis

The efficient markets hypothesis stipulates that all of the available information that is relevant to the price of an asset is already embodied in that price. It is based upon the argument that there is a large body of investors who react immediately (and at no cost to themselves) to any fresh information to buy or sell that asset . For example, if prices are expected to rise tomorrow, investors will buy today and in doing so, cause the price to rise until it is no longer expected to rise further. The existence in the market of noise traders need not invalidate the hypothesis, provided that most traders act rationally and that those who do not, make only random mistakes.

The formal statement that "in an informationally efficient market, price changes must be unforcastable if they are properly anticipated" was put forward and proved by Paul Samuelson in 1965 , and there followed a debate as to whether stock markets do in fact operate as efficient markets. That question was subsequently explored in studies undertaken and summarised by the economist Eugene Fama in 1991 . and others. Fama concluded that there is no important evidence to suggest that prices do not adjust to publicly available information.

A distinction has since been drawn between the hypothesis that prices are unforecastable (termed the "weak version") and the wider-ranging hypothesis that prices correctly reflect all of the relevant information (termed the "strong version"). The weak version has come to command acceptance as broad generalisation - although subject to rare exceptions (concerning the very few traders who have consistently outperformed the market). The strong version is less widely accepted because it implies the impossibility of speculative bubbles, and the futility of market regulation.

Financial instability hypothesis

The financial instability hypothesis, as formulated in 1992 by Hyman Minsky, states that the economy has financing regimes under which it is stable, and financing regimes in which it is unstable and that, during periods of sustained growth it tends to transform itself from stability to instability. It is based upon the argument that there are three categories of financial transaction:

• hedge transactions, whose payment obligations can all be met from their cash flows;
• speculative transactions for which interest payments can be met from their cash flows the repayment of principle has to be met from cash flows and are met by rolling over (ie the issue of more debt;.
• "Ponzi" transactions, for which neither interest payments nor capital repayments can be met from cash flows;

- and that the incidence of speculative and Ponzi transactions, that are responsible for the creation of bubbles, tend to increase in time of prosperity.

Financial models

The Capital Asset Pricing Model

The rate of return, r,  from an equity asset is given by

$r=r_{f}\beta (r_{m}-r_{f})$

where

r_f  is the risk-free rate of return

r_m  is the equity market rate of return

(and r_m - r_f is known as the equity risk premium)

and β is the covariance of the asset's return with market's return divided by the variance of the market's return.

(for a proof of this theorem see David Blake Financial Market Analysis page 297 McGraw Hill 1990)

The Arbitrage Pricing Model

The rate of return on the ith asset in a portfolio of n assets, subject to the influences of factors j=1 to k is given by

${r_{i}}={r_{f}}+\sum _{j=1}^{k}{y_{j}}{\beta _{ij}}$

where

${\beta _{ij}}={\frac {Cov(r_{i},d_{j})}{Var(d_{j})}}$

and

$y_{j}$ is the weighting multiple for factor $j$

$Cov(r_{i},d_{j})$ is the covariance between the return on the ith asset and the jth factor,

$Var(d_{j})$ is the variance of the jth factor

Black-Scholes option pricing model

The fair price,P, of a call option on a security is given by:

$P=CN(d_{1})-Xe^{-rt}N(d_{2})\,$

where:

C is the current price of the security;

$N(d_{1})$ is the cumulative probability distribution for the standard normal variate from -∞ to $d_{1}$;

X is the exercise price (see options definition);

r is the risk-free interest rate;

t is the time to expiry of the option;

$d_{1}$ and $d_{2}$ are given by the equations:

$d_{1}={\frac {\ln(C/X)+(r+\sigma ^{2}/2)t}{\sigma {\sqrt {t}}}}$;

$d_{2}=d_{1}-\sigma {\sqrt {t}}$;

and

$\sigma$ is the standard deviation (or volatility) of the price of the asset.

The first expression, $CN(d_{1})$,   of the equation is the expected benefit from acquiring a stock outright, obtained by multiplying the asset price by the change in the call premium with respect to a change in the underlying asset price. The second expression,  $Xe^{-rt}N(d_{2})$, is the present value of paying the exercise price on the expiration day. The fair market value of the call option is then calculated by taking the difference between these two parts

The underlying assumptions include:

• Dividend payments are not included;
• Options cannot be exercise before the stipulated date;
• Markets are efficient;
• No commissions are paid;
• Volatility is constant;
• The interest rate is constant; and,
• Returns are log-normally distributed.

Gambler's ruin

If q is the risk of losing one throw in a win-or-lose winner-takes-all game in which an amount c is repeatedly staked, and k is the amount with which the gambler starts, then the risk, r, of losing it all is given by:

r  =  (q/p)^(k/c)

where p  =  (1 - q),  and q  ≠  1/2

(for a fuller exposition, see Miller & Starr Executive Decisions and Operations Research Chapter 12, Prentice Hall 1960)