Hello,
Need your help,
My research topic is impact of dividend policy and profitability on share prices. I have taken 29 nonfinancial companies for 10 years time span. I ran simple OLS model in E-View 12 Student Version. But, I faced the problem of positive autocorrelation. I also applied panel data analysis - Pooled OLS, Fixed Effect, Random Effect Model. Is it require to fulfill the assumptions i. e. Normality of Residuals, Homoskedasticy of Residuals, No Autocorrelation, No Multicollinearity? E-View doesn't allow diagnostic tests in Panel Data analysis. I have no much idea about analysis and econometrics. Guide me. So, I can complete my multiple linear regression analysis.
Thank you in Advance.
For static panel data analysis, at first you need to test OLS. Then you will test LM test, if LM test does not suggest OLS. Then you need to test Fixed effect. However, you need to compare fixed effect by testing random effect. Finally, you need to test hausman test to decide the best result whether from fixed effect or random effect. If hausman test suggest fixed effect, then you will explain your research result based on the result generated from fixed effect test and vice versa @Shraddha G Raj
Thank you sir. I applied LM test and the p-value of Breusch-Pagan is 0.00, means Random Effect Model is better than POLS. Afterward, appplied redundant fixed effect test.P- value is 0.000, means FEM is better than POLS. Afterward, Hausman test is applied and the p-value is 0.000, means FEM is better. Still, I m in confusion about Assumptions...
It sounds as if your model is something like
share price_it = b0 + b1 dividend_policy_it + b2 profitability_it + u _it
for various definitions of b0. Before you do any econometrics I would suggest that you consider the following points
The idea is that before you go use your econometrics you use your knowledge of finance, economics, and common sense to set out the model which describes how the system works. When you have this model you can then test it. If it fails the tests either your data or your model is wrong.
The theory of share prices is very difficult. Very often random walk models outperform other models.
John C.frain sir. Yes sir, I selected my variables on the basis of theory and previous literature. I have no much detail idea about Econometrics because I am from Commerce background. I applied panel data analysis in Eview. I tested LM Breusch-Pagan test, Fixed Effect likelihood test and Hausman test for the selection of appropriate model. NFE model is suitable. I am in confusion.
I also am confused. Say your model is
share price_it = b0_i + b1_i dividend_policy_it + b2_i profitability_it + u _it
I have not used EVIEWS for many years and while it appears to be easy to use it may not be easy to replicate results. Please have a look at gretl
https://gretl.sourceforge.net/
which is free and, I think easier to use. Talk to your supervisor. He is in a better position to guide you. If you have no background in econometrics or statistics you may be trying to do too much. I stress that the underlying economic and finance theory must provide a model you can use in your estimation. Perhaps you might take a smaller collection of firms and do SUR estimation. Can I repeat my previous warning?
The theory of share prices is very difficult. Very often random walk models outperform other models.