Standardized beta value

[Raman SINGH]

Respected Sir

In one of your video on regression you taught that value of standardized beta can be calculated by converting the variables into z-value.
This is in the case of cross sectional data.
My question is whether this same technique is valid in panel data?
I have converted the variables into z-values and applied regession in e-views software, but ,results are quite different.
  So finally the question is how to get standardized beta value for panel data where fixed and random effect model is applied?

Thanking you

[Seema Malik]

I think there is no need to convert variables in to stand. Values in panel data.  The fixed effect model itself control the firm heterogenity.

I want to ask how did u calculate z score in panel data.  Inter firm or intra firm?

--
You received this message because you are subscribed to the Google Groups "DataAnalysis" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataanalysistrai...@googlegroups.com.
To post to this group, send email to dataanalys...@googlegroups.com.
Visit this group at http://groups.google.com/group/dataanalysistraining.
For more options, visit https://groups.google.com/d/optout.

[Raman SINGH]

Thanks for prompt reply
I calculated z value by calculating the mean and standard deviation of variable,then applying the formula (x-mean)/sigma to get z value..

What I want is standardized coefficient so that there would be a clear picture of impact of different independent variables on dependent variable.

Thanks

[Seema Malik]

I wanted to know whether u took mean of whole data set means total number of firm year observations and cal z score or  calculated z score of each firm separately.

[Raman SINGH]

Ok...
I got ur question.
I took the whole data set of that particular  variable (e.g. ROA) to calculate mean and accordingly calculated the z value of each firm specific.
Thank you

[Seema Malik]

According to my knowledge one need to standardize the data to make the variables unit free or to make it normal.
But there is no  need to do it in panel data. Standardisation and transformation both are different things. U can tranform your data in to log or any other form. But z score is not needed here.
I read some of the good papers on panel data.  Transformation was there but no standardisation.

[Raman SINGH]

Ok
U are talking about stationary of data, but, I am trying to calculate BETA (capital case) value which is called standardized coefficient value for a direct comparison of different independent variables on dependent variable respectively. This standardized coefficient makes the values unit free e.g. dollor, height etc. so that there is no problem of different units.
Now, the result will be in the form of standard variation unit ie. If there is one standard deviation change in independent variable then how much change will be there in dependent variable.
Usually in the papers only unstandardized coefficient value  beta (lower case)is given.
Thanks

[Preeti Saluja]

Recently I attended a workshop at IP university where the resource person has discussed that e views do not handle panel data.
For panel data use STATA. STATA will give you the within and between effects also.

[Raman SINGH]

Preeti

What you are saying is about cross sectional data which E-views doesn't handle when you want to use fixed effect model to measure the between and within effects. It will be handle by STATA software. 

The resource person never said that E-views doesn't handle panel data.(I was also there). It is the SPSS which is not suitable for panel data.  

I am not working on cross sectional data so there is no need of between and with in effect models. what i need is standardized coefficients value of fixed and random effect models by using E-views.

Thanks

[Seema Malik]

Totally agreed with raman ji.
Eviews can handle panel data. I also worked on panel data using eviews.
At that time I contacted so many experts and every body said that in panel data there is no need to apply z transformation. 

[Sonia Jindal]

if the data is panel in nature but variables does not unit free like no. of directors and remuneration of directors then the standardized coefficient is used or not used.