Likelihood estimation and inference in threshold regression
By Ping Yu. 2012
76 citations recorded by the source corpus.
This paper studies likelihood-based estimation and inference in parametric discontinuous threshold regression models with i.i.d. data. The setup allows heteroskedasticity and threshold effects in both mean and variance. By interpreting the threshold point as a “middle” boundary of the threshold variable, we find that the Bayes estimator is asymptotically efficient among all estimators in the locally asymptotically minimax sense. In particular, the Bayes estimator of the threshold point is asymptotically strictly more efficient than the left-endpoint maximum likelihood estimator and the newly proposed middle-point maximum likelihood estimator. Algorithms are developed to calculate asymptotic distributions and risk for the estimators of the threshold point. The posterior interval is proved to be an asymptotically valid confidence interval and is attractive in both length and coverage in finite samples.
Subfields: Statistics and Probability
View Paper: https://www.semanticscholar.org/paper/dd13a53104a4f1de857bd0538d45a68b5be394c0
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Statistics and Probability
Citations in the corpus
- Nested sub-sample search algorithm for estimation of threshold models (2016, methodology): “Usually they are obtained by the SGS algorithm; see, e.g., Caner (2002), Samia and Chan (2011), and Yu (2012).”
- Threshold Regression With a Threshold Boundary (2020, methodology): “Most literature uses the left-endpoint LSE (LLSE), while Yu (2012, 2015) shows that the middle-point LSE (MLSE) is more efficient in most cases.”
- Threshold Regression with Endogeneity (2017, unknown): “density fx|q(x|q) is bounded below and above for (x, q) ∈ X × Γε ; see Yu and Zhao (2013) for relaxing these conditions. The first part of Assumption F implies that there are no discrete covariates in x. As mentioned earlier in the remarks following Assumption S, this assumption ismade for simplicit”