Optimal Model Assessment, Selection, and Combination

By Xiaotong Shen, Hsin‐Cheng Huang. 2006

80 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/f1dd6d2c06c4fbb06900e986c88d3f8c8c847c8a

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

On the “degrees of freedom” of the lasso (2007, background): “Shen and Ye [20] and Shen, Huang and Ye [19] further proved that the degrees of freedom provides an adaptive model selection criterion that performs better than the fixed-penalty model selection criteria.”
Local Rademacher complexities and oracle inequalities in risk minimization (2006, background): “A general construction of covariance penalties can be found in [4, 7] in a family of Q-error losses including the Kullback–Leibler loss and the 0–1 classiﬁ-cation error loss.”
To Combine Forecasts or to Combine Information? (2010, methodology): “1 and Smith and Wallis (2008). (20)The MSFE ratios are computed with respect to the CI benchmark for Monte Carlo analysis (Section 4, Tables 1, 2), while they are computed with respect to the historical-mean benchmark for empirical analysis as in Campbell and Thompson (2008).”
Estimating a difference of Kullback–Leibler risks using a normalized difference of AIC (2008, methodology): “When the goal is prediction or estimating a parameter which may be common to several models, the model averaging approach (Hoeting et al., 1999; Hjort and Claesken, 2003; Shen and Huang, 2006) may be used.”
A unifying approach to the estimation of the conditional Akaike information in generalized linear mixed models (2014, methodology): “For instance the generalized SURE formula (Lemma 2) in Shen & Huang (2006) is a generalisation of Theorem 3.1 and Theorem 3.2 to distributions not necessarily from the exponential family.”