were on Total Operating Revenue, Total Assets and Total Equity from which the ratios on Return to Assets and Return to Equity were calculated using the following formulae -
Return to Assets = Total Operating Revenue / Total Assets
Return on Equity = Total Operating Revenue / Total Equity.
However, all the data obtained were not expressed in terms of a common currency. Thus, to arrive at a consistent result they were all converted in terms of US dollars.
After conversion two separate regressions, one for 2007 and the other for 2008, were run on the variables so obtained, with the Fortune Global 500 rankings of the individual companies as the dependent variable and ROA, ROE and EPS as the independent ones in each of them, since the purpose of the study is to find the existence of any significant relation between the stated dependent and the independent variables.
Methodology
The results so obtained give useful information on a number of aspects which are explained below.
Sign of Coefficients - They provide information on the relation that the variable holds with the dependent variable.
T-Statistics - They show the significance of relation with the dependent variable of each individual independent variable. An estimated t-statistic is compared with the tabulated one and the conclusions reached after a comparison are as follows :
If the estimated | t | > tabulated t at the given degrees of freedom and level of significance, we reject the null hypothesis at the particular level of significance.
If the estimated | t | < tabulated t at the given degrees of freedom and level of significance, we do not reject the null hypothesis at the particular level of significance.
F-Statistics - It shows how significantly all the independent variables can explain the variations in the dependent variable. According to the value of the estimated F-Statistic, we arrive at two conclusions as under :
If the estimated F > tabulated F at the given degrees of freedom and level of significance, we reject the null hypothesis at the particular level of significance.
If the estimated F < tabulated F at the given degrees of freedom and level of significance, we do not reject the null hypothesis at the particular level of significance.
R Square - This statistic helps to find out the goodness of fit of the estimated model. The regression that is used to estimate the behavior of the population is actually done on the basis of sample observations. This value of R Square shows how nicely the estimated model is can describe population variations.
With the help of all the above information, our next step will be to run the sample regression and analyse it properly so as to find how relevant is the relation between corporate social responsibility and financial performance of a firm.
Descriptive Statistics
The descriptive statistics of the variables being used are given as follows for both the years on whose data the analysis was carried out.
The regression results have been shown in the appendix to this chapter. We will explain them one by one on a yearly basis, i.e., first for the year 2007 and then for 2008. The regression has been carried on with the help of the statistical software STATA that gives informat
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。