On unification of the asymptotic theory of nonlinear econometric models

Hypothesis Testing with Efficient Method of Moments Estimation

“…The difference between this result and the type of result found by Burguete, Gallant, and Souza [1982], where the "likelihood ratio" test is not asymptotically chi-squared, is that we have based the test statistics on the optimal GMM estimator, for which WT= V-1. The difference between this result and the type of result found by Burguete, Gallant, and Souza [1982], where the "likelihood ratio" test is not asymptotically chi-squared, is that we have based the test statistics on the optimal GMM estimator, for which WT= V-1.…”

Productivity trends in U.S. manufacturing: Evidence from the NQ and AIM cost functions

“…; a n 2 k ; W 1 ; W 2 ; W 3 ; W 4 Þ, and wðp; y; t; hÞ is given by the right-hand side of (39)-(42). As Gallant and Golub (1984, p. 298) put it, all statistical estimation procedures that are commonly used in econometric research can be formulated as an optimization problem of the following type (Burguete et al, 1982) b h minimizes jðhÞ over Y…”

Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses

“…All the proofs are collected in the Appendix . Though there are more general assumptions on the true data generating process than Assumption Al (see, e.g., Gallant and Holly (1980 ), Burguette, Gallant and Souza (1982)), Assumption Al is the simplest assumption that still allows for the presence of exogenous variables . Following Vuong (1983), the vector I t is partitioned into No assumption is here made on the relationship between the two compe ting conditional models F e and G r in the sense that they may be nested, overlapping, or non-nested .…”