| Abstract |
The principal purpose of this research is to show theoretically that the generalised least squares estimators (GLSE) are more efficient than that of the odinary least squares estimators (OLSE) of the ...parameters for the specific form of heteroscedasticity that is, error variances are proportional to the square of the independent variable with the additional condition that sum of squares of the observations of the explanatong vaniable is equal to the number of observations that is，Σ ^ n _ i=1X ^ 2 _ i=n, where， “X” is the explanatory variable and “n” is the total number of observations. Moreover , the particular form of heteroscedasticity has been considered for a specific linear regression equation without the intercept term imposing the additional condition. In order to show theoretically the efficiency of the GLSE as compared to the OLSE of the parameters for the specific form of heteroscedasticity with the additional condition, we have to derive some important pr op er ties of the OLSE of parameter s for the specific form of heteroscedasticity with the additional condition that, Σ ^ n _ i=1X ^ 2 _ i=n
For empirical verification of the theoretical construct of this research, a set of real data on the total collections and collection of bound journals for 45 libraries in Bangladesh for the year 1995 has been used. The rationale behind the choice of the data set for the current purpose is also delineated.show more
|
Export Mendeley
104_p063