Percentage change in X variable instead of one-unit change

[Nadja Madsen]

Hi all

I am running a fixed effects model with independent variable, TotalFundingADA, and dependent variable, SBELAL3_1. My independent variable is a school's funding per student. While the "least-funded" schools get 33 dollars per student, the "most-funded" school gets 246.849 dollars per student. When running my fixed effects regression I get that a one-unit change (1 dollar) in TotalFundingADA leads to a .01 change in SBELAL3_1. SBELAL3_1 goes from 2200 to 2500, so it is a very small increase.



Is it possible to run a regression where I look at percentage increase in my independent variable. For example a 10 percent increase in funding leads to an .xx change in my dependent variable.

Thank you in advance!!!

[Nadja Madsen]

Perhaps a solution:
I have now simply created a new variable that has percentage increase per year for each unit, using:

tsset ID Year
gen fundingADA= D.TotalFundingADA/L.TotalFundingADA*100

And then I run my xtreg with this new variable, fundingADA.

[Phil Bromiley]

You didn't get a quick answer. You'll increase the chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. We also discourage pictures - as you can see, your results are hard to understand.

You can certainly do this the way you've programmed it. Note that your parameters will be the change in the change in y for a one unit change in the change in x.