Page 85 - Azerbaijan State University of Economics

THE JOURNAL OF ECONOMIC SCIENCES: THEORY AND PRACTICE, V.72, # 1, 2015, pp. 61-94

Where variables are following:

R(i) - rate of return of a security (i) over the period (t)

R (mt)- rate of return on market portfolio m

βi -the coefficient for the market portfolio return (expresses the sensitivity of a

security to financial market)

α – constant; and εit - the error term

The variable in formula suggest that variance of error term in time-series linear

regression same over the time.

Next step in our analysis is to estimate market model equation by ordinary least

squares (OLS) applying the data from estimation window. The use of OLS in market

model is justified by the fact that OLS is best linear unbiased (BLUE) estimate of

linear model coefficients when error are Gausian (Douglas and Simin (1999)). OLS

estimated coefficients for each security stock price in our sample are following

(MacKinlay, 1997):

Variance of error term:

Intercept of regression:

85