ingest cdrApp 2017-08-15T21:02:26.993Z d91e81c8-5a8a-4e8a-976c-cad4e396e5ee modifyDatastreamByValue RELS-EXT fedoraAdmin 2017-08-15T21:03:10.735Z Setting exclusive relation modifyDatastreamByValue RELS-EXT fedoraAdmin 2017-08-15T21:03:11.298Z Setting exclusive relation addDatastream MD_TECHNICAL fedoraAdmin 2017-08-15T21:03:11.899Z Adding technical metadata derived by FITS modifyDatastreamByValue RELS-EXT fedoraAdmin 2017-08-15T21:03:29.703Z Setting exclusive relation addDatastream MD_FULL_TEXT fedoraAdmin 2017-08-15T21:03:40.692Z Adding full text metadata extracted by Apache Tika modifyDatastreamByValue RELS-EXT fedoraAdmin 2017-08-15T21:03:58.526Z Setting exclusive relation modifyDatastreamByValue RELS-EXT cdrApp 2017-08-22T13:59:00.263Z Setting exclusive relation modifyDatastreamByValue MD_DESCRIPTIVE cdrApp 2018-01-25T18:48:43.161Z modifyDatastreamByValue MD_DESCRIPTIVE cdrApp 2018-01-27T18:23:12.214Z modifyDatastreamByValue MD_DESCRIPTIVE cdrApp 2018-03-14T16:14:16.841Z modifyDatastreamByValue MD_DESCRIPTIVE cdrApp 2018-05-18T18:34:59.258Z modifyDatastreamByValue MD_DESCRIPTIVE cdrApp 2018-07-11T14:55:31.582Z modifyDatastreamByValue MD_DESCRIPTIVE cdrApp 2018-07-18T10:34:00.746Z modifyDatastreamByValue MD_DESCRIPTIVE cdrApp 2018-08-21T19:21:58.262Z modifyDatastreamByValue MD_DESCRIPTIVE cdrApp 2018-09-27T20:11:17.099Z modifyDatastreamByValue MD_DESCRIPTIVE cdrApp 2018-10-12T10:41:20.563Z modifyDatastreamByValue MD_DESCRIPTIVE cdrApp 2018-10-17T15:59:38.754Z modifyDatastreamByValue MD_DESCRIPTIVE cdrApp 2019-03-21T20:58:59.700Z Rachel Nethery Author Department of Biostatistics Gillings School of Global Public Health Special Topics in Latent Variable Models with Spatially and Temporally Correlated Latent Variables The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. Spring 2017 2017 Biostatistics Neurosciences Social research EEG, Factor Analysis, Independent Component Analysis, Latent Variable Models, Social Vulnerability, Spatial Models eng Doctor of Philosophy Dissertation University of North Carolina at Chapel Hill Graduate School Degree granting institution Biostatistics Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text Rachel Nethery Creator Department of Biostatistics Gillings School of Global Public Health Special Topics in Latent Variable Models with Spatially and Temporally Correlated Latent Variables The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. Spring 2017 2017 Biostatistics Neurosciences Social research EEG, Factor Analysis, Independent Component Analysis, Latent Variable Models, Social Vulnerability, Spatial Models eng Doctor of Philosophy Dissertation University of North Carolina at Chapel Hill Graduate School Degree granting institution Biostatistics Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text Rachel Nethery Creator Department of Biostatistics Gillings School of Global Public Health Special Topics in Latent Variable Models with Spatially and Temporally Correlated Latent Variables The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. Spring 2017 2017 Biostatistics Neurosciences Social research EEG, Factor Analysis, Independent Component Analysis, Latent Variable Models, Social Vulnerability, Spatial Models eng Doctor of Philosophy Dissertation University of North Carolina at Chapel Hill Graduate School Degree granting institution Biostatistics Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text Rachel Nethery Creator Department of Biostatistics Gillings School of Global Public Health Special Topics in Latent Variable Models with Spatially and Temporally Correlated Latent Variables The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. 2017-05 2017 Biostatistics Neurosciences Social research EEG, Factor Analysis, Independent Component Analysis, Latent Variable Models, Social Vulnerability, Spatial Models eng Doctor of Philosophy Dissertation University of North Carolina at Chapel Hill Graduate School Degree granting institution Biostatistics Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text Rachel Nethery Creator Department of Biostatistics Gillings School of Global Public Health Special Topics in Latent Variable Models with Spatially and Temporally Correlated Latent Variables The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. 2017 Biostatistics Neurosciences Social research EEG, Factor Analysis, Independent Component Analysis, Latent Variable Models, Social Vulnerability, Spatial Models eng Doctor of Philosophy Dissertation University of North Carolina at Chapel Hill Graduate School Degree granting institution Biostatistics Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text 2017-05 Rachel Nethery Creator Department of Biostatistics Gillings School of Global Public Health Special Topics in Latent Variable Models with Spatially and Temporally Correlated Latent Variables The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. 2017 Biostatistics Neurosciences Social research EEG, Factor Analysis, Independent Component Analysis, Latent Variable Models, Social Vulnerability, Spatial Models eng Doctor of Philosophy Dissertation University of North Carolina at Chapel Hill Graduate School Degree granting institution Biostatistics Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text 2017-05 Rachel Nethery Creator Department of Biostatistics Gillings School of Global Public Health Special Topics in Latent Variable Models with Spatially and Temporally Correlated Latent Variables The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. 2017 Biostatistics Neurosciences Social research EEG, Factor Analysis, Independent Component Analysis, Latent Variable Models, Social Vulnerability, Spatial Models eng Doctor of Philosophy Dissertation University of North Carolina at Chapel Hill Graduate School Degree granting institution Biostatistics Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text 2017-05 Rachel Nethery Creator Department of Biostatistics Gillings School of Global Public Health Special Topics in Latent Variable Models with Spatially and Temporally Correlated Latent Variables The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. 2017 Biostatistics Neurosciences Social research EEG, Factor Analysis, Independent Component Analysis, Latent Variable Models, Social Vulnerability, Spatial Models eng Doctor of Philosophy Dissertation Biostatistics Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text 2017-05 University of North Carolina at Chapel Hill Degree granting institution Rachel Nethery Creator Department of Biostatistics Gillings School of Global Public Health Special Topics in Latent Variable Models with Spatially and Temporally Correlated Latent Variables The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. 2017 Biostatistics Neurosciences Social research EEG; Factor Analysis; Independent Component Analysis; Latent Variable Models; Social Vulnerability; Spatial Models eng Doctor of Philosophy Dissertation Biostatistics Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text 2017-05 University of North Carolina at Chapel Hill Degree granting institution Rachel Nethery Creator Department of Biostatistics Gillings School of Global Public Health Special Topics in Latent Variable Models with Spatially and Temporally Correlated Latent Variables The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. 2017 Biostatistics Neurosciences Social research EEG, Factor Analysis, Independent Component Analysis, Latent Variable Models, Social Vulnerability, Spatial Models eng Doctor of Philosophy Dissertation University of North Carolina at Chapel Hill Graduate School Degree granting institution Biostatistics Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text 2017-05 Rachel Nethery Creator Department of Biostatistics Gillings School of Global Public Health Special Topics in Latent Variable Models with Spatially and Temporally Correlated Latent Variables The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. 2017 Biostatistics Neurosciences Social research EEG, Factor Analysis, Independent Component Analysis, Latent Variable Models, Social Vulnerability, Spatial Models eng Doctor of Philosophy Dissertation Biostatistics Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text 2017-05 University of North Carolina at Chapel Hill Degree granting institution Rachel Nethery Creator Department of Biostatistics Gillings School of Global Public Health The term latent variable model (LVM) refers to any statistical procedure that utilizes information contained in a set of observed variables to construct a set of underlying latent variables that drive the observed values and associations. Independent component analysis (ICA) is a LVM that separates recorded mixtures of signals into independent source signals, called independent components (ICs). ICA is popular tool for separating brain signals of interest from artifacts and noise in electroencephalogram (EEG) data. Due to challenges in the estimation of uncertainties in ICA, standard errors are not generally estimated alongside ICA estimates and thus ICs representing brain signals of interest cannot be distinguished through a statistical hypothesis testing framework. In Chapter 2 of this dissertation, we propose a bootstrapping algorithm for ICA that produces bootstrap samples that retain critical correlation structures in the data. These are used to compute uncertainties for ICA parameter estimates and to construct a hypothesis test to identify ICs representing brain activity, which we demonstrate in the context of EEG functional connectivity. In Chapter 3, we extend this bootstrapping approach to accommodate pre-ICA dimension reduction procedures, and we use the resulting method to compare popular strategies for pre-ICA dimension reduction in EEG research. In the final chapter, we turn our attention to another LVM, factor analysis, which utilizes the covariance structure of a set of correlated observed variables to model a smaller number of unmeasured underlying variables. A spatial factor analysis (SFA) model can be used to quantify the social vulnerability of communities based on a set of observed social variables. Current SFA methodology is ill-equipped to handle spatial misalignment in the observed variables. We propose a joint spatial factor analysis model that identifies a common set of latent variables underlying spatially misaligned observed variables and produces results at the level of the smallest spatial units, thereby minimizing loss of information. We apply this model to spatially misaligned data to construct an index of community social vulnerability for Louisiana, which we integrate with Louisiana flood data to identify communities at high risk during natural disasters, based on both social and geographic features. 2017 Biostatistics Neurosciences Social research EEG; Factor Analysis; Independent Component Analysis; Latent Variable Models; Social Vulnerability; Spatial Models eng Doctor of Philosophy Dissertation Young Truong Thesis advisor Amy Herring Thesis advisor Alana Campbell Thesis advisor Eric Bair Thesis advisor Yun Li Thesis advisor text 2017-05 University of North Carolina at Chapel Hill Degree granting institution Nethery_unc_0153D_17166.pdf uuid:93a084a5-ca0f-43cb-beb4-cff1cfeb677d 2017-06-16T20:22:14Z 2019-08-15T00:00:00 proquest application/pdf 9216889 yes