Functional Forms of Independent Variables

[Prateek Bedi]

Hi,

If our objective is to ascertain the relationship (specifically, sign and significance of Beta coefficient) between independent variables and dependent variable in an OLS regression (cross sectional or time series or panel regressions), should we actually care about an appropriate functional form of the independent variables? I have observed that in order to establish such relationship and to explain variation in the dependent variable, research papers keep on suggesting fresh set of variables as potential determinants of dependent variable without discussing the issue of functional form. For example, in majority of corporate finance research, some financial variables (explanatory variables) are picked up from the financial statements to explain variation in another financial variable (dependent variable) without even making a mention of the issue of functional form.

However, I have read in Gujarati (2009) that one must transform an independent variable by looking at its graphical relationship with dependent variable and then include it in the model. Further, I also have not come across a research paper which reports the results of RAMSEY RESET to suggest the appropriateness of the functional forms used by them in the model.

Thanks!

P.S.: Cross posted at https://stats.stackexchange.com/ques...dent-variables

[Marcos Almeida]

There are several threads on a similar matter. You may start by taking a look at this thread.

I am not saying this is a necessary approach, I mean, transforming the IVs, but you may check its behaviour after a transformation with the commands - ladder - , - qladder - and - gladder - for that matter.

[Eddie Oczkowski]

The RESET is useful for indicating whther a functional form is appropriate if the null hypothesis of no specification is not rejected . However if the null hyothesis is rejected then it is non-informative, it does not tell you what functional form is appropriate.

The orginal Ramsey paper which developed the RESET test has been cited over 3000 times in goggle scholar, so you should be able to find many examples of its use.

[Prateek Bedi]

There are several threads on a similar matter. You may start by taking a look at this thread.

I am not saying this is a necessary approach, I mean, transforming the IVs, but you may check its behaviour after a transformation with the commands - ladder - , - qladder - and - gladder - for that matter.

Firstly, thanks a lot for your response. I checked this thread but could not find a discussion relevant for my question. Could you please identify the specific post which addresses my query?

Thanks!