ion automatically on the data being processed. Those information are sufficient to decide the acceptance or rejection of the corresponding null hypothesis.
In 2007, it is found that none of the explanatory variables can play a significant role in explaining variations in the dependent variable. It is evident from the p-value of the individual estimated coefficients. P-value shows the probability of Type-I error and thus, higher the p-value is, the less significant is the variable. In fact, we take a standard measure to take an decisions regarding the null hypotheses. Usually, we standardize a level of significance, which is nothing but the rejection area in the cumulative distribution curve and reject the null hypothesis if the p-value is greater than this level of significance. In most cases, this standard is considered to be 5%. So, our criterion in this case will be,
If p-value > 0.05, we do not reject the null hypothesis.
If p-value < 0.05, we reject the null hypothesis.
The above criterion also applies for testing the F-statistic, meant to test the significance of all the explanatory variables in explaining the variations in the dependent variable.
We also could have used the criterion mentioned in the methodology part and arrived at similar results. However, as the former process is more tedious, so we will stick to the one mentioned above.
T-statistic
For the year 2007, it is found that the t-statistic of each and every explanatory variable has a p-value greater than 0.05. So, it cannot be said that any of them play a vital role in explaining the variations in the CSR performance, indicated by their CSR rankings, significantly. Again for the year 2008, too, though the figures are different, but their implications are similar. None of the t-statistics for the variables display a significant p-value, implying that they fail to produce significant effects on the dependent variable. Thus in case of both years, for all the variables, the null hypothesis that each of them do not play a significant role in explaining variations in the dependent variable is ruled out. Speaking more statistically, in each case, the null hypotheses are accepted at 5% level of significance.
F-Statistic
The F-statistic for the year 2007, overall, however indicates a significant relation between the explanatory variables and the dependent variable. The p-value of the F-statistic is 0.0257, which is lower than 0.05 (5%), and hence implies that all the explanatory variables together can explain variations in the dependent variable, CSR rankings. But, on the contrary, the results for the year 2008 are exactly the reverse and imply no such significant relation. Thus, although in the first case, we do not reject the null hypothesis, in the latter one, we do reject it at 5% level of significance.
Adjusted R Square
In case of the regression run on the data for the year 2007, the adjusted R Square value is 0.4347, while for regression carried on using the 2008 year data, the value is 0.2148. it is obvious that neither of the values of the Adjusted R Square statistics are evident of the fact that the model being fitted is a good one. For a model to become a good estimate of the observed values, a rule of thumb is that the value of the Adjusted R Square statistic must be greater than 0.7. But, here both of them are found to
本论文由英语论文网提供整理,提供论文代写,英语论文代写,代写论文,代写英语论文,代写留学生论文,代写英文论文,留学生论文代写相关核心关键词搜索。