Multi-criteria optimization in regression

In this paper, we consider standard as well as instrumental variables regression. Specification problems related to autocorrelation, heteroskedasticity, neglected non-linearity, unsatisfactory out-of-small performance and endogeneity can be addressed in the context of multi-criteria optimization. The new technique performs well, it minimizes all these problems simultaneously, and eliminates them for the most part. Markov Chain Monte Carlo techniques are used to perform the computations. An empirical application to NASDAQ returns is provided.

Automated summary

In this paper, a new technique for regression analysis is introduced, which can effectively address common problems such as autocorrelation, heteroskedasticity, etc. through multi-criteria optimization, eliminating these problems for the most part.