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

Spatial Signal Detection Using Conti...

Jhuang, An-Ting.

FindBook

Google Book

Amazon

博客來

Spatial Signal Detection Using Continuous Shrinkage Priors.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Spatial Signal Detection Using Continuous Shrinkage Priors./
作者:	Jhuang, An-Ting.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2018,
面頁冊數:	119 p.
附註:	Source: Dissertations Abstracts International, Volume: 80-05, Section: B.
Contained By:	Dissertations Abstracts International80-05B.
標題:	Biostatistics. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=11007122
ISBN:	9780438598164

Spatial Signal Detection Using Continuous Shrinkage Priors.
Jhuang, An-Ting.

Spatial Signal Detection Using Continuous Shrinkage Priors. - Ann Arbor : ProQuest Dissertations & Theses, 2018 - 119 p.

Source: Dissertations Abstracts International, Volume: 80-05, Section: B.

Thesis (Ph.D.)--North Carolina State University, 2018.

This item must not be sold to any third party vendors.

The statistical problem of identifying spatial regions affected by an experimental factor has many applications, including epidemiology, neuroscience and materials science. Motivated by the problem of detecting changes in two-dimensional X-ray diffraction data, we propose a Bayesian spatial model for sparse signal detection in Chapter 2. Our model places considerable mass near zero and has heavy tails to reflect the prior belief that the image signal is zero for most pixels and large for an important subset. We show that the spatial prior places mass on nearby locations simultaneously being zero, and also allows for nearby locations to simultaneously be large signals. The form of the prior facilitates efficient computing for large images. We conduct a simulation study to evaluate the properties of the proposed prior and show that it outperforms other spatial models. We apply our method in the analysis of X-ray diffraction data from a two-dimensional area detector to detect changes in the pattern when the material is exposed to an electric field. In Chapter 3 we extend the spatial horseshoe prior to the spatiotemporal setting. Periodontitis is a chronic in ammatory disease that affects the gum tissue and bone supporting the teeth. The whole-mouth average pocket depth has been commonly used as an indicator of periodontal diagnosis, without considering the spatial correlation across teeth. The objective is to flag local regions of deterioration for individual patients simultaneously. We achieve the objective via the spatial horseshoe prior on spatially-varying linear-trend coefficients. We also consider a low-rank representation to capture the nonstationary covariance structure of the PD data and reduce the computational burden so the method can be applied to thousands of subjects. The empirical results exhibit an improvement in prediction when we use shrinkage priors with nonstationary covariance. Exposure uncertainty is a common concern in health studies such as the periodontal example of Chapter 3. As an example, in Chapter 4 we consider maternal exposure to ambient air pollution during pregnancy has been linked with adverse birth outcomes in recent studies. However, there are remaining uncertainties and challenges in understanding the effects of particulate matter exposure on neonatal health including uncertainty of exposure measurement errors using environmental monitoring data, the variation in sources of particulate matter, the identification of a critical window of exposure, and the residential mobility. Our access to geocoded addresses for the pregnant women gives us an opportunity to examine potential effects of air pollution on fetal development. In Chapter 4, we investigate the misclassification effect of fine particulate matter due to residential mobility on evaluating the risk of congenital heart defects. Consideration of maternal residence during pregnancy does not lead to significant difference in the empirical results. However, the simulation study shows that the estimation bias and mean squared errors increase as the moving proportion is higher and the moving distance is longer.

ISBN: 9780438598164Subjects--Topical Terms:

1002712
Biostatistics.
Subjects--Index Terms:

Bayesian variable selection

Spatial Signal Detection Using Continuous Shrinkage Priors.
LDR:04429nmm a2200373 4500 001 2280347
005 20210827095929.5
008 220723s2018 ||||||||||||||||| ||eng d
020 $a 9780438598164
035 $a (MiAaPQ)AAI11007122
035 $a AAI11007122
040 $a MiAaPQ $c MiAaPQ
100 1 $a Jhuang, An-Ting. $3 3558860
245 1 0 $a Spatial Signal Detection Using Continuous Shrinkage Priors.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2018
300 $a 119 p.
500 $a Source: Dissertations Abstracts International, Volume: 80-05, Section: B.
500 $a Publisher info.: Dissertation/Thesis.
502 $a Thesis (Ph.D.)--North Carolina State University, 2018.
506 $a This item must not be sold to any third party vendors.
506 $a This item must not be added to any third party search indexes.
520 $a The statistical problem of identifying spatial regions affected by an experimental factor has many applications, including epidemiology, neuroscience and materials science. Motivated by the problem of detecting changes in two-dimensional X-ray diffraction data, we propose a Bayesian spatial model for sparse signal detection in Chapter 2. Our model places considerable mass near zero and has heavy tails to reflect the prior belief that the image signal is zero for most pixels and large for an important subset. We show that the spatial prior places mass on nearby locations simultaneously being zero, and also allows for nearby locations to simultaneously be large signals. The form of the prior facilitates efficient computing for large images. We conduct a simulation study to evaluate the properties of the proposed prior and show that it outperforms other spatial models. We apply our method in the analysis of X-ray diffraction data from a two-dimensional area detector to detect changes in the pattern when the material is exposed to an electric field. In Chapter 3 we extend the spatial horseshoe prior to the spatiotemporal setting. Periodontitis is a chronic in ammatory disease that affects the gum tissue and bone supporting the teeth. The whole-mouth average pocket depth has been commonly used as an indicator of periodontal diagnosis, without considering the spatial correlation across teeth. The objective is to flag local regions of deterioration for individual patients simultaneously. We achieve the objective via the spatial horseshoe prior on spatially-varying linear-trend coefficients. We also consider a low-rank representation to capture the nonstationary covariance structure of the PD data and reduce the computational burden so the method can be applied to thousands of subjects. The empirical results exhibit an improvement in prediction when we use shrinkage priors with nonstationary covariance. Exposure uncertainty is a common concern in health studies such as the periodontal example of Chapter 3. As an example, in Chapter 4 we consider maternal exposure to ambient air pollution during pregnancy has been linked with adverse birth outcomes in recent studies. However, there are remaining uncertainties and challenges in understanding the effects of particulate matter exposure on neonatal health including uncertainty of exposure measurement errors using environmental monitoring data, the variation in sources of particulate matter, the identification of a critical window of exposure, and the residential mobility. Our access to geocoded addresses for the pregnant women gives us an opportunity to examine potential effects of air pollution on fetal development. In Chapter 4, we investigate the misclassification effect of fine particulate matter due to residential mobility on evaluating the risk of congenital heart defects. Consideration of maternal residence during pregnancy does not lead to significant difference in the empirical results. However, the simulation study shows that the estimation bias and mean squared errors increase as the moving proportion is higher and the moving distance is longer.
590 $a School code: 0155.
650 4 $a Biostatistics. $3 1002712
650 4 $a Statistics. $3 517247
653 $a Bayesian variable selection
653 $a Exposure misclassification
653 $a Functional data analysis
653 $a High-dimensional data
653 $a Space-time disease surveillance
690 $a 0308
690 $a 0463
710 2 $a North Carolina State University. $b Statistics. $3 2098676
773 0 $t Dissertations Abstracts International $g 80-05B.
790 $a 0155
791 $a Ph.D.
792 $a 2018
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=11007122