ON VARIABLE SELECTION IN LINEAR REGRESSION

By Paul Kabaila. Econometric Theory, 2002

105 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Molecular Biology, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/67d34a543e134ea157f8effdcde11b319e0c849f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Molecular Biology

In Statistics and Probability

Citations in the corpus

Sparse estimators and the oracle property, or the return of Hodges’ estimator (2007, unknown): “…of a linear regression model, but inspection of the proof shows that it easily extends far beyond that 3That pointwise asymptotics can be misleading in the context of model selection has been noted earlier in Hosoya (1984), Shibata (1986a), Pötscher (1991), and Kabaila (1995, 2002). framework.”
Parametric or nonparametric? A parametricness index for model selection (2011, unknown): “(see, e.g., [21, 40, 44]).”
How Anticipated Employee Mobility Affects Acquisition Likelihood: Evidence from a Natural Experiment (2015, methodology): “In Column 3, we use backwards elimination (Lindsey and Sheather, 2010) to remove controls that reduce an optimal Bayesian Information Criterion and find results that are consistent with the Full Model, suggesting that our findings are not sensitive to the Copyright © 2014 John Wiley & Sons, Ltd.…”
Model Selection (2018, background): “…problem is that the asymptotic loss-efficiency results discussed above are only pointwise results and do not hold uniformly w.r.t. the parameter θ: Kabaila (2002), shows, in a simple setting where (21) holds, that for given sample size n, there exists a parameter θ = θn with infinitely many…”
Variable Selection in Predictive Regressions (2013, background): “Kabaila (2002) cautions that result in Shibata (1981) that favors the AIC over the BIC is based on first fixing the data generating process, and then providing a pointwise analysis of β as T increases.”
Least squares model averaging by Mallows criterion (2010, methodology): “For example, Kabaila (2002) demonstrated that post-model-selection estimators with asymptotically efficient properties can have rather inefficient small sample performance.”