3
$\begingroup$

I am using OLS model to predict overall price of fixed phone montly subscription in order to estimate price index. Explanatory variables are only the one user looks after when signing phone contract like free minutes, megabytes etc. Among these variables I've included level of call price as something that user buys as any other features that form a bundle = phone plan.

It was argued that call price is causing endogeneity.

Mine explanation is that the possible bias in the coefficients should not be exaggerated; they're are not as important in estimating overall price of phone plan to calculate price index as in estimating e.g. price elasticity of demand.

What could be a reasonable explanation to rebut this claim about endogeneity? Any other ideas?

$\endgroup$
5
  • $\begingroup$ You can't "explain away" endogeneity. You have to demonstrate that it's not an issue if anyone raised it. $\endgroup$ – Aksakal Dec 31 '14 at 16:31
  • $\begingroup$ @Aksakal In what way? $\endgroup$ – Quirik Dec 31 '14 at 16:32
  • $\begingroup$ such as Granger causality tests. When you can't rule it out, you'll have to deal with it either with instrumental variables or by using lags. $\endgroup$ – Aksakal Dec 31 '14 at 17:17
  • $\begingroup$ What exactly was the argument for why call price is causing endogeneity? It would be easier to evaluate, and possibly rebut, this argument if you also write down the equation you are estimating and specify which variables you want to give a causal interpretation. $\endgroup$ – CloseToC Dec 31 '14 at 17:31
  • $\begingroup$ It was argued by my supervisor that the price is likely to cause endogeneity not only because of potential colinearity with other variables but also because of reverse causality. I'm thinking that this assumption is based on previous research on model for estimating price elasticity of demand. As I said, the purpose of my model is to estimate overall price of phone plan based on its characteristics to calculate price index. $\endgroup$ – Quirik Dec 31 '14 at 18:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.