Assignment in

Quantitative methods of investment

analysis

Submitted to

Satya Priya

Submitted by

MD.Mayaze
ID: 170421006
Batch 1 (7TH semester)
 Quantitative methods of investment analysis

Statistical concepts applying on the quantitative methods of investment analysis. So those

Investment income and risk

> Return on investment and expected rate of return:

Investment risk & Variance and standard deviation.

> Relationship between risk and return:

Covariance

Correlation and Coefficient of determination.

>Relationship between the returns on asset and market portfolio:

The characteristic line and the Beta factor.

Investment income and risk:

A return is the ultimate objective for any investor. But a relationship between return and risk is a
key concept in finance. As finance and investments areas are built upon a common set of
financial principles, the main characteristics of any investment are investment return and risk.
However, to compare various alternatives of investments the precise quantitative measures for
both of these characteristics are needed.

Return on investment and expected rate of return

General definition of return is the benefit associated with an investment. In most cases the
investor can estimate his/ her historical return precisely.

Many investments have two components of their measurable return:

> a capital gain or loss;

> some form of income;

Rate of return = Income + Capital gains / Purchase price (%)

For example, rate of return of the share (r) will be estimated:

formula 1,

Here

D - dividends;

Pmb - market price of stock at the beginning of holding period;

Pme - market price of stock at the end of the holding period.

The rate of return, calculated in formulas 2.2 and 2.3 is called holding period return, because its
calculation is independent of the passages of the time. All the investor knows is that there is a
beginning of the investment period and an end.

Investor can’t compare the alternative investments using holding period returns, if their holding
periods (investment periods) are different. Statistical data which can be used for the investment
analysis and portfolio formation deals with a series of holding period returns. For example,
investor knows monthly returns for a year of two stocks.

How he/ she can compare these series of returns? In these cases, arithmetic average return or
sample mean of the returns (ř) can be used:

Formula 2,

Here

ri - rate of return in period i;

n - number of observations.
But both holding period returns and sample mean of returns are calculated using historical data.
However, what happened in the past for the investor is not as important as what happens in the
future, because all the investors ‘decisions are focused to the future, or to expected results from
the investments.

Analyzing each particular investment vehicle possibilities to earn income in the future investor
must think about several „scenarios “of probable changes in macro economy, industry and
company which could influence asset prices ant rate of return. Theoretically it could be a series
of discrete possible rates of return in the future for the same asset with the different probabilities
of earning the particular rate of return. But for the same asset the sum of all probabilities of these
rates of returns must be equal to 1 or 100 %. In mathematical statistics it is called simple
probability distribution.

The expected rate of return(r) of investment is the statistical measure of return, which is the sum
of all possible rates of returns for the same investment weighted by probabilities:

Here

hi - probability of rate of return;

ri - rate of return.

In all cases than investor has enough information for modeling of future scenarios of changes in
rate of return for investment, the decisions should be based on estimated expected rate of return.

ample mean can give an unbiased estimate of the expected value, but obviously it‘s not perfectly
accurate, because based on the assumption that the returns in the future will be the same as in the
past.
In general, the sample mean of returns should be taken for as long time, as investor is confident
there has not been significant change in the shape of historical rate of return probability
distribution.

Investment risk:

Risk can be defined as a chance that the actual outcome from an investment will differ from the
expected outcome. Obvious, that most investors are concerned that the actual outcome will be
less than the expected outcome. The more variable the possible outcomes that can occur, the
greater the risk. Risk is associated with the dispersion in the likely outcome. And dispersion
refers to variability. So, the total risk of investments can be measured with such common
absolute measures used in statistics as

• variance; • standard deviation.

Variance can be calculated as a potential deviation of each possible investment rate of return
from the expected rate of return:

Formula 3,

To compute the variance in formula 3 all the rates of returns which were observed in estimating
expected rate of return (ri) have to be taken together with their probabilities of appearance (hi).

The other an equivalent to variance measure of the total risk is standard deviation which is
calculated as the square root of the variance:
Formula 4,

In the cases than the arithmetic average return or sample

Sample standard deviation (δr) consequently can be calculated as the square root of the sample
variance: formula 6,

Variance and standard deviation are used when investor is focused on estimating total risk that
could be expected in the defined period in the future. Sample variance and sample standard
deviation are more often used when investor evaluates total risk of his /her investments during
historical period – this is important in investment portfolio management.

Relationship between risk and return

The expected rate of return and the variance or standard deviation provide investor with
information about the nature of the probability distribution associated with a single asset.

But how does one asset having some specific trade-off between return and risk influence the
other one with the different characteristics of return and risk in the same portfolio? And what
could be the influence of this relationship to the investor’s portfolio?
The statistics that can provide the investor with the information to answer these questions are
covariance and correlation coefficient. Covariance and correlation are related and they generally
measure the same phenomenon – the relationship between two variables. Both concepts are best

understood by looking at the math behind them.

Covariance

Two methods of covariance estimation can be used: the sample covariance and the population
covariance. The sample covariance is estimated than the investor has n’t enough information
about the underlying probability distributions for the returns of two assets and then the sample of
historical returns is used. Sample covariance between two assets - A and B is defined in the next
formula (7):

here rA,t , rB,t - consequently, rate of return for assets A and B in the time period t, when t varies
from 1 to n; ŕA, ŕB -sample mean of rate of returns for assets A and B consequently.

As can be understood from the formula, a number of sample covariance can range from “–” to
“+” infinity. Though, the covariance number doesn’t tell the investor much about the relationship
between the returns on the two assets if only this pair of assets in the portfolio is analyzed. It is
difficult to conclude if the relationship between returns of two assets (A and B) is strong or
weak, taking into account the absolute number of the sample variance.

Positive number of covariance shows that rates of return of two assets are moving to the same
direction: when return on asset A is above its mean of return (positive), the other asset B is
tending to be the same (positive) and vice versa: when the rate of return of asset A is negative or
below its mean of return, the returns of other asset tend to be negative too. Negative number of
covariance shows that rates of return of two assets are moving in the contrariwise directions:
when return on asset A is above its mean of return (positive), the returns of the other asset - B is
tend to be the negative and vice versa. Though, in analyzing relationship between the assets in
the same portfolio using covariance for portfolio formation it is important to identify which of
the three possible outcomes exists:

> positive covariance (“+”),

> negative covariance (“-”) or zero covariance (“0”).

If the positive covariance between two assets is identified the common recommendation for the
investor would be not to put both of these assets to the same portfolio, because their returns
move in the same direction and the risk in portfolio will be not diversified.

If the zero covariance between two assets is identified it means that there is no relationship
between the rates of return of two assets. The assets could be included in the same portfolio, but
it is rare case in practice and usually covariance tends to be positive or negative.

When the historical rates of return of two assets known for the investor are marked in the area
formed by axes ŕA, ŕB, it is very easy to identify what kind of relationship between two assets
exists simply by calculating the number of observations in each:

if the number of observations in sections I and III prevails over the number of observations in
sections II and IV, the covariance between two assets is positive (“+”);

if the number of observations in sections II and IV prevails over the number of observations in
sections I and III, the covariance between two assets is negative (“-”);

if the number of observations in sections I and III equals the number of observations in sections
II and IV, there is the zero covariance between two assets (“0”).
The population covariance is estimated when the investor has enough information about the
underlying probability distributions for the returns of two assets and can identify the actual
probabilities of various pairs of the returns for two assets at the same time.

Similar to using the sample covariance, in the population covariance case the graphical method
can be used for the identification of the direction of the relationship between two assets. But the
graphical presentation of data in this case is more complicated because three dimensions must be
used (including the probability).

Correlation and Coefficient of determination

Correlation is the degree of relationship between two variables.

The correlation coefficient between two assets is closely related to their covariance. The
correlation coefficient between two assets A and B (kAB) can be calculated using the next
formula:
here δ (rA) and δ(rB)are standard deviation for asset A and B consequently.

Very important, that instead of covariance when the calculated number is unbounded, the
correlation coefficient can range only from -1,0 to +1,0. The closer the absolute meaning of the
correlation coefficient to 1,0, the stronger the relationship between the returns of two assets. Two

variables are perfectly positively correlated if correlation coefficient is +1,0, that means that the
returns of two assets have a perfect positive linear relationship to each other (see Fig. 2.4), and
perfectly negatively correlated if correlation coefficient is -1,0, that means the asset returns have
a perfect inverse linear relationship to each other (see Fig. 2.5). But most often correlation
between assets returns is imperfect (see Fig. 2.6). When correlation coefficient equals 0, there is
no linear relationship between the returns on the two assets (see Fig. 2.7). Combining two assets
with zero correlation with each other reduces the risk of the portfolio. While a zero correlation
between two assets returns is better than positive correlation, it does not provide the risk

reduction results of a negative correlation coefficient.

Therefore, as it was pointed out earlier, the covariance primarily provides information to the
investor about whether the relationship between asset returns is positive, negative or zero,
because simply observing the number itself without any context with which to compare the
number, is not very useful. When the covariance is positive, the correlation coefficient will be
also positive, when the covariance is negative, the correlation coefficient will be also negative.
But using correlation coefficients instead of covariance investor can immediately asses the
degree of relationship between assets returns.

The characteristic line and the Beta factor

Before examining the relationship between a specific asset and the market portfolio the concept
of “market portfolio” needs to be defined. Theoretical interpretation of the market portfolio is
that it involves every single risky asset in the global economic system, and contains each asset in
proportion to the total market value of that asset relative to the total value of all other assets
(value weighted portfolio). But going from conceptual to practical approach - how to measure
the return of the market portfolio in such a broad its understanding - the market index for this
purpose can be used. Investors can think of the market portfolio as the ultimate market index.
And if the investor following his/her investment policy makes the decision to invest, for
example, only in stocks, the market portfolio practically can be presented by one of the available
representative indexes in particular stock exchange. The most often the relationship between the
asset return and market portfolio return is demonstrated and examined using the common stocks
as assets, but the same concept can be used analyzing bonds, or any other assets. With the given
historical data about the returns on the particular common stock (rJ) and market index return
(rM) in the same periods of time investor can draw the stock’s characteristic line (see Fig. 2.8.).

Stock’s characteristic line: describes the relationship between the stock and the market; shows
the return investor expect the stock to produce, given that a particular rate of return appears for
the market; helps to assess the risk characteristics of one stock relative to the market. Stock’s
characteristic line as a straight line can be described by its slope and by point in which it crosses
the vertical axis - intercept (point A in Fig. 2.8.). The slope of the characteristic line is called the
Beta factor. Beta factor for the stock J and can be calculated using following formula: a

here: Cov(rJ,rM) – covariance between returns of stock J and the market portfolio; δ²(rM) -
variance of returns on market portfolio. The Beta factor of the stock is an indicator of the degree
to which the stock reacts to the changes in the returns of the market portfolio. The Beta gives the
answer to the investor how much the stock return will change when the market return will
change by 1 percent.

Intercept AJ (the point where characteristic line passes through the vertical axis) can be
calculated using following formula:

AJ = rJ - ββββJ×××× rM,

here:

rJ- rate of return of stock J;

βJ - Beta factor for the stock J;

rM - rate of return of the market.

The intercept technically is a convenient point for drawing a characteristic line. The
interpretation of the intercept from the investor’s point of view is that it shows what would be the
rate of return of the stock, if the rate of return in the market is zero.

Summary

1. The main characteristics of any investment are investment return and risk. However, to
compare various alternatives of investments the precise quantitative measures for both of
these characteristics are needed.
2. General definition of return is the benefit associated with an investment. Many
investments have two components of their measurable return: (1) a capital gain or loss;
(2) some form of income. The holding period return is the percentage increase in returns
associated with the holding period.
3. Risk can be defined as a chance that the actual outcome from an investment will differ
from the expected outcome. The total risk of investments can be measured with such
common absolute measures used in statistics as variance and standard deviation. Variance
can be calculated as a potential deviation of each possible investment rate of return from
the expected rate of return. Standard deviation is calculated as the square root of the
variance. The more variable the possible outcomes that can occur, the greater the risk.
4. Covariance and correlation coefficient are used to answer the question, what is the
relationship between the returns on different assets. Covariance and correlation
coefficient are related and they generally measure the same phenomenon – the
relationship between two variables.
5. Analyzing relationship between the assets in the same portfolio using covariance for
portfolio formation it is important to identify which of the three possible outcomes exists:
positive covariance, negative covariance or zero covariance.