The subtle danger of symmetry restrictions in time series regressions, with application to fertility models.

Abstract

It is widely known that linear restrictions involve bias. What is not known is that some linear restrictions are especially dangerous for hypothesis testing. For some, the expected value of the restricted coefficient does not lie between (among) the true unconstrained coefficients, which implies that the estimate is not a simple average of these… (More)

Topics

Figures and Tables

Sorry, we couldn't extract any figures or tables for this paper.

Slides referencing similar topics