SAS - REGRESSION ANALYSIS 2

Final Exam

1. Run the OLS regression of q (market to book value of assets) on gindex (the total number of anti-takeover provisions adopted by a firm, also called “G-Index” or “Governance-Index”), after controlling for all control variables used in a similar manner as in our prior class.

2. Comment on the SAS regression output including

a. Coefficient estimates on the explanatory variables (the sign and magnitude that indicates the economic significance) and statistical significance (t-stat & p-value)

The Board has a great p value in LN_MV of 16.43 which is greater than 5%. Which means that the company has an increase in their natural logarithm of market value of equity. Moreover, their LEV_AT is -5.38 which is less than .05. That shows If the ratio is less than 0.5, most of the company's assets are financed through equity and that have a positive impact on the companies economy.If p-value is <.0001 we will conclude that mean is statistically significantly different from zero. Therefore, the Board’s LN_MV and LEV_AT are significantly different from zero.

b. Overall fit and validity of the model: F-stat and its p-value

c. Goodness of fit of the model: R2 and Adjusted R2.

R-squared is a statistical measure of how approximately the data are to the fitted regression line. The value of R squared is 0.3671 while the value of Adj. R squared is 0.3629. A linear model clarifies the percentage of the response variable variation, so higher R-square the better the model fits your data because 100 percent of R-square the more fitted around it’s mean, but 0% of R-square which does not explain any of the variation in the response variable around its mean. the Board’s R-square is 36.71% ,in this case the variance in stock return can be predicted. Which is not too low, that means that the model is fit for this data.

3. Run the IV/2SLS regression of q on gindex and the control variables, where using industry average gindex (“gindex_mean”) as the instrument for gindex.

4. What is the endogeneity problem you face here? Explain the sources/reasons for the endogeneity.

5. Comment on the IV/2SLS regression output including coefficient estimates on the explanatory variables (the sign and magnitude that indicates the economic significance) and statistical significance (t-stat & p-value).

6. Comment on the Endogeneity test result using the PORC QLIM procedure (at the end of the output).

7. Compare the OLS and IV/2SLS coefficient estimates & statistical significance using either the PROC SYSLIN or PROC QLIM procedure.

2.2 If you were the referee for a paper that examines the relationship between various board of directors characteristics and firms’ innovation output (e.g., patents, new products, R&D). After controlling for various observable macro, industry and year effects, firm financial, accounting factors, executive characteristics, board characteristics etc. (i.e., controlled for anything you can think of that are relevant), the authors find that busy directors have a negative and significant impact on firms’ innovation output. Busy directors are defined as the percentage of directors on a firm’s board that hold at least three (>=3) outside directorships on other firms’ boards. The authors interpret their findings as being consistent with the notion that busy directors, whose time and efforts are in limited supply, are stretched out than non-busy directors and thus are lax in monitoring and advising firms’ innovation activity.

a) To establish the causality from busy directors to innovation, what is the econometric issue the authors should address?

b) How would you design your study to address the issue – be as specific as possible?

c) What instrument(s) would you use if you were conducting this study and using instrumental variables estimation to address the problem?