東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Missing data methods in marine mamma...

Shotwell, Mary E.

Linked to FindBook

Google Book

Amazon

博客來

Missing data methods in marine mammal strandings research.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Missing data methods in marine mammal strandings research./
Author:	Shotwell, Mary E.
Description:	178 p.
Notes:	Source: Dissertation Abstracts International, Volume: 72-06, Section: B, page: .
Contained By:	Dissertation Abstracts International72-06B.
Subject:	Statistics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3451918
ISBN:	9781124575087

Missing data methods in marine mammal strandings research.
Shotwell, Mary E.

Missing data methods in marine mammal strandings research. - 178 p.

Source: Dissertation Abstracts International, Volume: 72-06, Section: B, page: .

Thesis (Ph.D.)--Medical University of South Carolina, 2010.

The completeness of marine mammal strandings data has been an issue since the start of the Marine Mammal Health and Stranding Response Program (MMHSRP). Missing data are inherent on many levels of stranding sampling. Only those that have been stranded (vs. wild) are sampled. All stranded animals are either detected or go undetected. Out of those animals that have been stranded and detected, some are fully measured and others are not. Measurement completeness can depend on both unforeseen circumstances (weather, accessibility, volunteer availability), and systematic circumstances stemming from the decreased feasibility of transporting larger (vs. smaller) animals for comprehensive laboratory measurements.

ISBN: 9781124575087Subjects--Topical Terms:

517247
Statistics.

Missing data methods in marine mammal strandings research.
LDR:02700nam 2200289 4500 001 1402513
005 20111102140018.5
008 130515s2010 ||||||||||||||||| ||eng d
020 $a 9781124575087
035 $a (UMI)AAI3451918
035 $a AAI3451918
040 $a UMI $c UMI
100 1 $a Shotwell, Mary E. $3 1681707
245 1 0 $a Missing data methods in marine mammal strandings research.
300 $a 178 p.
500 $a Source: Dissertation Abstracts International, Volume: 72-06, Section: B, page: .
500 $a Adviser: Elizabeth Slate.
502 $a Thesis (Ph.D.)--Medical University of South Carolina, 2010.
520 $a The completeness of marine mammal strandings data has been an issue since the start of the Marine Mammal Health and Stranding Response Program (MMHSRP). Missing data are inherent on many levels of stranding sampling. Only those that have been stranded (vs. wild) are sampled. All stranded animals are either detected or go undetected. Out of those animals that have been stranded and detected, some are fully measured and others are not. Measurement completeness can depend on both unforeseen circumstances (weather, accessibility, volunteer availability), and systematic circumstances stemming from the decreased feasibility of transporting larger (vs. smaller) animals for comprehensive laboratory measurements.
520 $a A statistical model will be built piecewise to account for biases introduced at two levels of missingness, design and implement simulation studies for evaluating the impact of each model component, and apply the resulting methods to determining growth as a function of total body weight using South Carolina strandings data. A Bayesian approach will be used in a mixture model setting that weights observed data and unobserved data to account for selection bias. The weight of the unobserved data component in the mixture relies on total body length (measure of size) and a random component, both of which are suspected reasons for missingness. Model performance under misspecification will be evaluated. A pseudo-data approach will be adapted to a Bayesian context to account for detection bias. Both simple and complex detection bias will be investigated. The final model will combine both pieces into one unifying Bayesian model to be applied to bottlenose dolphin (Tursiops truncatus) strandings data from South Carolina collected to estimate total body weight growth with reduced bias.
590 $a School code: 0122.
650 4 $a Statistics. $3 517247
650 4 $a Biology, Bioinformatics. $3 1018415
690 $a 0463
690 $a 0715
710 2 $a Medical University of South Carolina. $3 700119
773 0 $t Dissertation Abstracts International $g 72-06B.
790 1 0 $a Slate, Elizabeth, $e advisor
790 $a 0122
791 $a Ph.D.
792 $a 2010
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3451918