英语论文网

above, and the second part will adds some controlling variables to the model 1 to form model 2 and explore further the application of CAPM in the real management practice, and the controlling variables include market value growth rate, asset growth rate and the debt growth rate of the company.

Table 1 Model Summary without controlling variables 

Model R R Square Adjusted R Square Std. Error of the Estimate
1 0.272a 0.074 0.065 1.00497
a. Predictors: (Constant), x 

Table 1 indicates the result of regression fitting of model 1, the correlation index is 0.272, which means that the variables have positive relationships. However, the R Square of the regression fitting is only 0.074, which means that the relationships between the explaining and explained variables are not significant enough. 

Table 2 ANOVA Analysis without controlling variables 
Model Sum of Squares df Mean Square F Sig.
1 Regression 2.239 1 2.239 2.217 0.137a
Residual 434.288 430 1.010 
Total 436.528 431 
a. Predictors: (Constant), x 
b. Dependent Variable: y 

Table 2 is the test result for the whole formula, as we could find in the table, the F-test value is 2.217, and the P is 0.137, which means that the regression model is significant as a whole, but the F-test result is not satisfying, that is also to say, some regularities as for the distribution of the ROE can’t be explained by the explaining variable-Beta. 

Table 3 Coefficients Analysis without controlling variables 
Model Unstandardized Coefficients Standardized Coefficients t Sig.
B Std. Error Beta 
1 (Constant) 2.828 0.238 11.878 0.000
X -0.177 -0.119 -0.072 -1.489 0.137
a. Dependent Variable: y 

Table 3 gives out the T-test for the regression model. T-test explores the significance of the parameters which are estimated in the regression. As table 3 tells, the test value of constant and the coefficient of X are 11.878 and -1.489 respectively, and the P value are 0 and 0.137 accordingly. The T-test value and P value of the coefficients help us to get the conclusion that all the coefficients are significant in the regression model.
2. The analysis with the controlling variables embodied in the model
The empirical model above has a serious problem that is the missing of controlling variables. As we all know that, there are many factors which could influence the ROE of the companies, and the variables such as basic interest rate and return rate of market portfolio can’t cover all these factors. In that case, it is necessary for us to add some suitable explaining variables into the empirical model and enforce the explaining power of the model. 
Here we choose market value of the company, the total assets and total debts as the controlling variables for the ROE of the company, and we carry out the ln transformation for these variables, so the mv, assets and debts in the following tables represent ln(mv), ln(assets) and ln(debts). Model 2 is exhibited as:
(3)
Here xi represents the variables such as , ln(mv), ln(assets) and ln(debts) and the explained variable y becomes .

Table 4 Model Summary for the model with controlling variables

Model R R Square Adjusted R Square Std. Error of the Estimate
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。