ABSTRACT
We propose a two-part model to deal with the problem of taking logarithm when the independent variable in a regression contains zeros. In particular, we generalize the conventional approach by choosing the optimal constant to add to the variable before taking logarithm, treating observations with zero values as a separate group. The empirical evidence supports our proposed model.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 The scaled IHST in Carroll et al. (Citation2003), , can be viewed as GIHST plus a constant.
2 See of the Appendix for the complete results.
3 See of the Appendix for the complete results.
4 Although we focus on the log–log model, a similar method is applicable to the semi-log model as well.