2004
DOI: 10.3150/bj/1093265631
|View full text |Cite
|
Sign up to set email alerts
|

Semiparametric density estimation under a two-sample density ratio model

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

2
41
0
1

Year Published

2009
2009
2024
2024

Publication Types

Select...
6
2
1

Relationship

0
9

Authors

Journals

citations
Cited by 60 publications
(44 citation statements)
references
References 13 publications
2
41
0
1
Order By: Relevance
“…However, this naive approach is not reliable in high-dimensional problems since division by an estimated quantity can magnify the estimation error of the dividend. To overcome this drawback, various approaches to directly estimating density-ratios without going through density estimation have been explored recently, including the moment matching approach (Gretton et al 2009), the probabilistic classification approach (Qin 1998;Cheng and Chu 2004), the density matching approach Tsuboi et al 2009;Yamada and Sugiyama 2009;Nguyen et al 2010;, and the density-ratio fitting approach ).…”
Section: Introductionmentioning
confidence: 99%
“…However, this naive approach is not reliable in high-dimensional problems since division by an estimated quantity can magnify the estimation error of the dividend. To overcome this drawback, various approaches to directly estimating density-ratios without going through density estimation have been explored recently, including the moment matching approach (Gretton et al 2009), the probabilistic classification approach (Qin 1998;Cheng and Chu 2004), the density matching approach Tsuboi et al 2009;Yamada and Sugiyama 2009;Nguyen et al 2010;, and the density-ratio fitting approach ).…”
Section: Introductionmentioning
confidence: 99%
“…2.3). The kernel mean matching method [14] and the logistic regression based method [3], [6], [30] also allow one to directly estimate a density ratio q(x)/q (x). However, the derivation of these methods heavily relies on the fact that the two density functions q(x) and q (x) share the same domain, which is not fulfilled in the current setting.…”
Section: Other Methods Of Density Ratio Estimationmentioning
confidence: 99%
“…We first give a comprehensive review of methods for estimating the density ratio, including the kernel density estimator 18) , kernel mean matching 4) , logistic regression 6), 19), 20) , the Kullback-Leibler importance estimation procedure 9), 10) , least-squares importance fitting 11) , and unconstrained least-squares importance fitting 11) . Note that "importance" refers to the density ratio, derived from importance sampling 21) .…”
Section: Introductionmentioning
confidence: 99%