I have data where the dependent variable are counts of an event. I am modeling the relationship between the dependent and independent variables using a negative binomial model, but I was also hoping to try some machine learning or nonparametric models, specifically ones that will handle nonlinearity in the responses and predictors. Any suggestions? Thanks.
$\begingroup$
$\endgroup$
2
-
1$\begingroup$ What kind of nonparametric do you mean? Not depending on some specific probability model for the counts, like Poisson or negative binomial? Or not using some specific functional form for the predictors? In the last case, loook into splines and specifically gam's (generalized additive models). In R, the package mgcv. $\endgroup$– kjetil b halvorsen ♦Commented Feb 19, 2016 at 13:42
-
$\begingroup$ Thanks. I want to reduce parametric and model dependence, so I'm looking for a very flexible method to model the response and predictors. The gams show that I do have nonlinear relationships, so I wanted to offer a further robustness check. $\endgroup$– Rick726Commented Feb 19, 2016 at 14:09
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
In Cameron and Trivedi (2013) Chapter 11.6 is a section about nonparametric methods for count data. I haven't used nonparametric methods for count data sofar but it seems that most standard nonparametric methods such as kernel methods, nearest neighbor or spline regression are also available for count data.
You can find also some information in the np-package for R, see Chapter 4.
If you are mostly interested in getting consistent coefficients, than your results might be quite robust to the detailed functional form as long as the conditional mean is correctly specified (see Quasi Maximum Likelihood Methods, see again Cameron and Trivedi).
But probably the easiest way is to check if you need nonparametric methids is to use quantile regression methods for count data which seem to be available for Stata.
answered Feb 19, 2016 at 14:02