To address the bias caused by the lagged dependent variable we employ a quasi-maximum likelihood (QML) estimator, which is asymptotically equivalent to, but has advantages over, the widely used system GMM estimation.
Specifically we implement this quasi-maximum likelihood (QML) estimator via the ST ATA command developed by (Kripfganz, 2016).
Ruiz, "Quasi-maximum likelihood
estimation of stochastic volatility models," Journal of Econometrics, vol.
However, research on other topics has demonstrated that that method can produce inconsistent estimates of the percentage differences in the expected values of wages, whereas quasi-maximum likelihood methods provide consistent estimates.
All of those studies suggested using quasi-maximum likelihood estimators (QMLEs) with the exponential form of the model, which leaves the dependent variable untransformed.
Papke and Wooldridge proposed a quasi-maximum likelihood
estimator (QMLE) of [beta].
But the discussion of quasi-maximum likelihood
estimators in this model has been scarcely seen.
The estimation carried out by the quasi-maximum likelihood
method is consistent and asymptotically follows a normal distribution.
With these evaluations available, the door for likelihood-based inference is open, either by searching for a maximum of the function(Quasi-Maximum Likelihood
estimation) or by simulating the posterior distribution of the parameters using a Markov Chain Monte Carlo algorithm(Bayesian estimation).
[phi] and [mu] are quasi-maximum likelihood
estimates of the corresponding parameters, N is the number of observations, p is the number of parameters (usually N = n(n - 1) and p = 2n - 1), while S[E.sub.b]([mu]) is the bootstrap estimate of the standard error of the estimator [mu], i.e.,
The author uses a quasi-maximum likelihood
estimation procedure of Robinson [QMLE, 1995a], which has some advantages with respect to other methods.
(10.) We computed the quasi-maximum likelihood
covariances and standard errors as described in Bollerslev and Wooldridge (1992).