東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Missing data imputation for tree-bas...

He, Yan.

FindBook

Google Book

Amazon

博客來

Missing data imputation for tree-based models.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Missing data imputation for tree-based models./
作者:	He, Yan.
面頁冊數:	81 p.
附註:	Adviser: Richard Berk.
Contained By:	Dissertation Abstracts International67-07B.
標題:	Statistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3226020
ISBN:	9780542796487

Missing data imputation for tree-based models.
He, Yan.

Missing data imputation for tree-based models. - 81 p.

Adviser: Richard Berk.

Thesis (Ph.D.)--University of California, Los Angeles, 2006.

A wide variety of data can include some form of censorship or missing information. Missing data are a problem for all statistical analyses, tree-based models, such as CART and Random Forests are certainly no exception.

ISBN: 9780542796487Subjects--Topical Terms:

517247
Statistics.

Missing data imputation for tree-based models.
LDR:02662nam 2200301 a 45 001 968847
005 20110920
008 110921s2006 eng d
020 $a 9780542796487
035 $a (UnM)AAI3226020
035 $a AAI3226020
040 $a UnM $c UnM
100 1 $a He, Yan. $3 1292904
245 1 0 $a Missing data imputation for tree-based models.
300 $a 81 p.
500 $a Adviser: Richard Berk.
500 $a Source: Dissertation Abstracts International, Volume: 67-07, Section: B, page: 3876.
502 $a Thesis (Ph.D.)--University of California, Los Angeles, 2006.
520 $a A wide variety of data can include some form of censorship or missing information. Missing data are a problem for all statistical analyses, tree-based models, such as CART and Random Forests are certainly no exception.
520 $a In recent years, there have been many new developed tools that can be applied to missing data problems: likelihood and estimating function methodology, cross-validation, the bootstrap and other simulation techniques, Bayesian and multiple imputations, and the EM algorithm. Although applied successfully to well-defined parametric models, such methods may be inappropriate for tree-based models, which are usually considered as non-parametric models. CART/RF have built-in algorithms to impute missing data, such as surrogate variables or proximity. But these imputation methods have no formal rationale, and are unstable, especially for RF models.
520 $a The nonparametric bootstrap methods to impute missing values overcome all of the drawbacks that are implicit in both single and multiple imputations. It (1) does not depend on the missing-data mechanism, (2) requires no knowledge of either the probability distributions or model structure, and (3) successfully incorporates the estimates of uncertainty associated with the imputed data. Furthermore, 2000 replications of bootstrap samples provide stable and accurate statistical inferences (Efron, 1994).
520 $a In my dissertation research, the nonparametric bootstrap methods were implemented to impute missing values before cases were dropped down the tree (CART/RF), and the classification results were compared to both complete-data/full-data analysis and to the classification results using surrogate variables/proximity. Significant improvement in the ability to predict were found for both CART and RF models.
590 $a School code: 0031.
650 4 $a Statistics. $3 517247
690 $a 0463
710 2 0 $a University of California, Los Angeles. $3 626622
773 0 $t Dissertation Abstracts International $g 67-07B.
790 $a 0031
790 1 0 $a Berk, Richard, $e advisor
791 $a Ph.D.
792 $a 2006
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3226020