Mathematics and

Statistics for
Finance

By: Manoj Aryan

 Variance
Variance: The occurrence of an event may deviate from the mean or an expected value.
The spread of such occurrences around the expected value can be measured by variance.
Thus, variance equals to average of squares of the deviation of each value from the mean.
It may be expressed as:
 Standard Deviation
Standard deviation has been used as a proxy measure for risk of a security.
It measures the fluctuations around mean returns. It equals to the positive
square root of variance.

Where it is necessary to distinguish the standard deviation of a population from the standard
deviation of a sample drawn from this population, we often use the symbol ‘s’ for the letter
and ‘s ’(lower case Greek sigma) for the former. Thus, ‘s 2 ’and ‘ s 2 ’ would represent the
sample variance and population variance, respectively. Sometimes the standard deviation of a
sample’s data is defined with (N-1) replacing N in the denominations in above equations
because the resulting value represents a better estimate for the standard deviation of a
population from which the sample is taken. For large values of N (certainly N>30), there is
practically no difference between the two definitions.
 Coefficient of Variation
The actual dispersion/variation as determined by standard deviation, is called absolute
dispersion. Relative dispersion, on the other hand gives feel about absolute dispersion relative
to mean/average. In other words, If the absolute dispersion is the standard deviation (s ) and
average is the mean ( X ), then the relative dispersion is called ‘coefficient of dispersion’ or
‘coefficient of variation’ (V). It is given by:

It is generally expressed as a percentage.

Example 6: Security A gives a return of 12% with a dispersion of 4%, while security B gives a
return of 20% with a dispersion of 5%. Which security is more risky?
Coefficient of Variation for Security A = (4/12) = 0.33 or 33% and
Coefficient of Variation for Security B = (5/20) = 0.25 or 25%. Therefore, the security A is
more risky in relation to its return.
 Covariance
Covariance describes the nature of relationship between two variables/securities. If X and
Y are two securities, then the covariance between the two securities is given by the
following formula:

When two securities are combined, if rates of return of two securities move together, their
interactive risk/covariance is said to be positive and vice versa. If rates of return are
independent, then the covariance is zero.
Coefficient of Correlation
Coefficient of correlation is another measure designed to indicate the similarity or
dissimilarity in the behavior of two variables (here two securities x and y). The total
variation consists of explained variation as well as unexplained variation. The ratio of
the explained variation to the total variation is called the ‘co-efficient of
determination’. Since the ratio is always non-negative, it is denoted by r2 .The quantity
xy r is called the coefficient of correlation and is given by:
 Risk
Risk may be described as variability/fluctuation/deviation of actual return from
expected return from a given asset/investment. Higher the variability, greater is the risk.
In other words, the more certain the return from an asset, lesser is the variability and
thereby lesser is the risk.

Types of Risks
The risk of a security can be broadly classified into two types such as systematic risk
and unsystematic risk. Standard deviation has been used as a proxy measure for total
risk.

Unsystematic risk

Systematic Risk
 Systematic Risk
Systematic Risk refers to that portion of total variability (/risk) in return caused by factors
affecting the prices of all securities. Economic, political, and sociological changes are the
main sources of systematic risk. Though it affects all the securities in the market, the
extend to which it affects a security will vary from one security to another. Systematic risk
can not be diversified. Systematic risk can be measured in terms of Beta (ß), a statistical
measure. The beta for market portfolio is equal to one by definition. Beta of one (ß=1),
indicates that volatility of return on the security is same as the market or index; beta more
than one (ß>1) indicates that the security has more unavoidable risk or is more volatile than
market as a whole, and beta less than one (ß<1) indicates that the security has less
systematic risk or is less volatile than market.
 Unsystematic risk

Unsystematic Risk refers to that portion of total risk that is unique or peculiar to
a firm or an industry, above and beyond that affecting securities markets in
general. Factors like consumer preferences, labour strikes, management
capability etc. cause unsystematic risk (/variability of returns) for a company’s
stock. Unlike systematic risk, the unsystematic risk can be reduced/avoided
through diversification. Total risk of a fully diversified portfolio equals to the
market risk of the portfolio as its specific risk becomes zero.
 Return and Risk of a
Single Asset

For example, for a security if price at the beginning of the year is Rs. 50.00; dividend
receivable at the end of the year is Rs. 2.50; and price at the end of the year is Rs. 55.00
then, the rate of return on this security is:
 Return of a Portfolio
Investors prefer investing in a portfolio of assets (combination of two or more
securities/assets) rather than investing in a single asset. The expected returns on a portfolio
is a weighted average of the expected returns of individual securities or assets comprising
the portfolio. The weights are equal to the proportion to amount invested in each security to
the total amount.
 Risk of a Portfolio
According to the Modern Portfolio Theory, while the expected return of a portfolio is a
weighted average of the expected returns of individual securities (or assets) included in
the portfolio, the risk of a portfolio measured by variance (or standard deviation) is not
equal to the weighted average of the risk of individual securities included in the
portfolio. The risk of a portfolio not only depends on variance/risk of individual
securities but also on co-variances between the returns on the individual securities.
 Capital Asset Pricing Model
(CAPM)
Portfolio Theory developed by Harry Markowitz is essentially a normative approach as it
prescribes what a rational investor should do. On the other hand, Capital Asset Pricing
Model (CAPM) developed by William Sharpe and others is an exercise in positive
economics as it is concerned with (i) what is the relationship between risk and return for
efficient portfolio? and (ii) What is the relationship between risk and return for an
individual security? CAPM assumes that individuals are risk averse. CAPM describes the
relationship/trade-off between risk and expected/required return. It explains the behavior
of security prices and provides mechanism to assess the impact of an investment in a
proposed security on risks and return of investors’ overall portfolio. The CAPM provides
framework for understanding the basic risk-return trade-offs involved in various types of
investment decisions. It enables drawing certain implications about risk and the size of
risk premiums necessary to compensate for bearing risks. Using beta (ß) as the measure of
non-diversifiable risk, the CAPM is used to define the required return on a security
according to the following equation:

Rs =Rf + ßs (Rm - Rf )
Where: Rs – required return; Rf – Risk free return; ßs – Beta factor of the security; Rm
– Market return of the security
Finish