Statistical methodology for environmental applications /
We consider the problem of estimating fetal growth and length of gestation for bowhead whales, Balaena mysticetus, of the Bering, Chukchi, Beaufort Seas stock. This western arctic population is carefully monitored by the International Whaling Commission and the Alaska Eskimo Whaling Commission. Quot...
| Main Author: | |
|---|---|
| Format: | Thesis Book |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
1999.
|
| Subjects: | |
| Online Access: | http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=731681441&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD |
| Summary: | We consider the problem of estimating fetal growth and length of gestation for bowhead whales, Balaena mysticetus, of the Bering, Chukchi, Beaufort Seas stock. This western arctic population is carefully monitored by the International Whaling Commission and the Alaska Eskimo Whaling Commission. Quota determination is assisted by biological information, such as fetal growth and gestation, which is part of a population dynamics model used to estimate the annual replacement yield of the stock. We use a Bayesian hierarchical nonlinear model for fetal growth and provide predictive distributions for both gestation length and conception dates. These results are also used to propose estimates of geographic locations for both conception and parturition. Tidal variation is an environmental application of interest to not only scientists, but in courts of law as well. Prediction of water levels in the Gulf of Mexico was the topic of a multi-million dollar lawsuit against the state of Texas in 1993. The water level data used in the law suit contains a substantial amount of missing data. In addition, certain meteorological predictors have a substantial amount of missing data. One requirement of the state was to produce predicted values for the water level and any predictors used in modeling. This research presents predictions of the missing water levels and covariates based on space-time covariance structures. We also present a spatial regression model based on the observed and predicted values. We propose models for the measurement error incurred when using predicted values instead of observed values and examine properties of the spatial regression coefficients based on the predicted response and predictor variables. |
|---|---|
| Item Description: | Vita. "Major Subject: Statistics". |
| Physical Description: | xii, 113 leaves : illustrations ; 28 cm. Issued also on microfiche from University Microfilm Inc. |
| Bibliography: | Includes bibliographical references (leaves 95-102). |