Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Interpretation of interaction term in log linear (non linear) model

From   David Hoaglin <>
Subject   Re: st: Interpretation of interaction term in log linear (non linear) model
Date   Fri, 21 Jun 2013 23:16:50 -0400

Dear Suryadipta,

Thank you for the compliment.  As it happens, though, I have not had
enough information about your data and your analysis to write anything
approaching a referee's report.

Assessing the fit of a model should involve much more than calculating
a single number such as R-squared.  One usually looks for influential
data and makes a variety of plots of residuals.  If the papers that
have used Poisson models used data that had a substantial percentage
of zeros, and did not do anything special about the zeros, I suspect
that those models did not fit very well.  Did the authors not give any
empirical evidence on how well their models fit?

If propensity to import could be treated as a binary outcome (positive
imports versus zero imports), one part of the analysis could use
logistic regression (or a probit model, if you prefer).

If theory favors the use of a fixed effect for each trading pair, what
does that theory say about how to handle the connections among trading
pairs that involve the same country?

The project in which we split each set of data into three parts
produced several papers.  The first of those papers is
Pine M, Jordan HS, Elixhauser A, et al.  Enhancement of claims data to
improve risk adjustment of hospital mortality.  Journal of the
American Medical Association 2007; 297:71-76.
The chapter on model assessment and selection in the book by Hastie,
Tibshirani, and Friedman (2009) has some discussion of splitting a
dataset into a training set (50%), a validation set (25%), and a test
set (25%).

David Hoaglin

Hastie T, Tibshirani R, Friedman J (2009).  The Elements of
Statistical Learning.  Springer.

On Tue, Jun 18, 2013 at 6:05 PM, Suryadipta Roy <> wrote:
> Dear David,
> Thank you very much for the comments and the wonderful suggestions!
> These almost read like a referee report! I had to take some time to
> reply to your comments since the issues that you have raised are
> substantive. Theoretical work in the gravity model of trade literature
> mainly suggest the importance of structural factors that prevent
> countries from trading with each other. Some of the important papers
> have used -heckman- selection models but that model is more suitable
> to explain why countries export (or do not export), while my research
> question focuses on the propensity to import. Moreover, the exclusion
> restrictions in the selection equation are still not very well
> founded. The papers that have used Poisson models have not reported
> the goodness of fit. -poisson- by itself reports a pseudo-rsquare
> which is not comparable to the linear r-square, while -xtpoisson- or
> -xtpqml- that I have implemented does not report any r-square. The
> cluster-robust standard errors address both the problems of
> overdispersion and serial correlation (Cameron and Trivedi,
> Microeconometrics using Stata, 2010). It is theory here that guides
> the use of fixed effects, e.g. I need to incorporate 5638 trading pair
> fixed effects (since I have 76 countries with complete data, I can
> have a maximum of 76*75 = 5700 trading pair relationships).
> Your suggestions on model building by splitting the data have been
> extremely illuminating. However, I was wondering if you could give me
> a bit more concrete suggestions as to how to go about it, e.g. could
> you kindly give me the reference to the paper where you undertook the
> data splitting approach so that I could read a bit more about it?
> Best regards,
> Suryadipta.
*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index