Re: st: Non-linear regression

[Daniel Feenberg <feenberg@nber.org>]

On Tue, 8 Feb 2011, Hamizah Hassan wrote:

> Greetings all,
>
> Firstly, I apologize because my question might be more towards 
> statistics rather than Stata. But hopefully there will be a statistician 
> out there that could help me. I would like to run non-linear regression 
> by including the linear and quadratic functions of the variable. I just 
> realize that if the variable is in percentage, the quadratic figure is 
> higher than the linear figure. However, if it is in decimal, it would be 
> the other way around and definitely it will effect on the meaning of the 
> results. Could anybody clarify on this matter?

It is an elementary question, but still a good one. The answer is that the 
values of the squared percentage are always going to be 10,000 (100 
squared) times as great as the values of the squared percentage. So the 
least squares estimate of the coefficient will always be 10,000 times 
smaller, and none of the other coefficients need to change, nor do any of 
the summary statistics such as R**2.

Daniel Feenberg

> Thanks in advance.
>
> Hamizah Hassan
> RMIT University
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/