Philip Reiss - Academia.edu (original) (raw)
Papers by Philip Reiss
Journal of Computational and Graphical Statistics a Joint Publication of American Statistical Association Institute of Mathematical Statistics Interface Foundation of North America, Jul 1, 2012
In linear regression with functional predictors and scalar responses, it may be advantageous, par... more In linear regression with functional predictors and scalar responses, it may be advantageous, particularly if the function is thought to contain features at many scales, to restrict the coefficient function to the span of a wavelet basis, thereby converting the problem into one of variable selection. If the coefficient function is sparsely represented in the wavelet domain, we may employ the well-known LASSO to select a relatively small number of nonzero wavelet coefficients. This is a natural approach to take but to date, the properties of such an estimator have not been studied. In this paper we describe the wavelet-based LASSO approach to regressing scalars on functions and investigate both its asymptotic convergence and its finite-sample performance through both simulation and real-data application. We compare the performance of this approach with existing methods and find that the wavelet-based LASSO performs relatively well, particularly when the true coefficient function is spiky. Source code to implement the method and data sets used in the study are provided as supplemental materials available online.
Journal of Computational and Graphical Statistics, Feb 12, 2014
Http Dx Doi Org 10 1198 016214507000000527, 2012
ABSTRACT
Journal of the American Statistical Association, Feb 1, 2006
Biostatistics (Oxford, England), Jan 29, 2016
SummaryMany modern neuroimaging studies acquire large spatial images of the brain observed sequen... more SummaryMany modern neuroimaging studies acquire large spatial images of the brain observed sequentially over time. Such data are often stored in the forms of matrices. To model these matrix-variate data we introduce a class of separable processes using explicit latent process modeling. To account for the size and two-way structure of the data, we extend principal component analysis to achieve dimensionality reduction at the individual level. We introduce necessary identifiability conditions for each model and develop scalable estimation procedures. The method is motivated by and applied to a functional magnetic resonance imaging study designed to analyze the relationship between pain and brain activity.
Journal of child psychology and psychiatry, and allied disciplines, Jan 22, 2016
Social anxiety disorder (SAD) typically onsets in adolescence and is associated with multiple imp... more Social anxiety disorder (SAD) typically onsets in adolescence and is associated with multiple impairments. Despite promising clinical interventions, most socially anxious adolescents remain untreated. To address this clinical neglect, we developed a school-based, 12-week group intervention for youth with SAD, Skills for Academic and Social Success (SASS). When implemented by psychologists, SASS has been found effective. To promote dissemination and optimize treatment access, we tested whether school counselors could be effective treatment providers. We randomized 138, ninth through 11th graders with SAD to one of three conditions: (a) SASS delivered by school counselors (C-SASS), (b) SASS delivered by psychologists (P-SASS), or (c) a control condition, Skills for Life (SFL), a nonspecific counseling program. Blind, independent, evaluations were conducted with parents and adolescents at baseline, post-intervention, and 5 months beyond treatment completion. We hypothesized that C-SASS...
Modern data-rich analyses may call for fitting a large number of nonparametric quantile regressio... more Modern data-rich analyses may call for fitting a large number of nonparametric quantile regressions. For example, growth charts may be constructed for each of a collection of variables, to identify those for which individuals with a disorder tend to fall in the tails of their age-specific distribution; such variables might serve as developmental biomarkers. When such a large set of analyses are carried out by penalized spline smoothing, reliable automatic selection of the smoothing parameter is particularly important. We show that two popular methods for smoothness selection may tend to overfit when estimating extreme quantiles as a smooth function of a predictor such as age; and that improved results can be obtained by multifold cross-validation or by a novel likelihood approach. A simulation study, and an application to a functional magnetic resonance imaging data set, demonstrate the favorable performance of our methods.
Journal of the American Academy of Child & Adolescent Psychiatry, 2015
Diffusion tensor imaging (DTI) can identify structural connectivity alterations in attention-defi... more Diffusion tensor imaging (DTI) can identify structural connectivity alterations in attention-deficit/hyperactivity disorder (ADHD). Most ADHD DTI studies have concentrated on regional differences in fractional anisotropy (FA) despite its limited sensitivity to complex white matter architecture and increasing evidence of global brain differences in ADHD. Here, we examine multiple DTI metrics in separate samples of children and adults with and without ADHD with a principal focus on global between-group differences. Two samples: adults with ADHD (n = 42) and without (n = 65) and children with ADHD (n = 82) and without (n = 80) were separately group matched for age, sex, and head motion. Five DTI metrics (FA, axial diffusivity, radial diffusivity, mean diffusivity, and mode of anisotropy) were analyzed via tract-based spatial statistics. Group analyses tested for diagnostic differences at the global (averaged across the entire white matter skeleton) and regional level for each metric. Robust global group differences in diffusion indices were found in adults, with the largest effect size for mode of anisotropy (MA; Cohen's d = 1.45). Global MA also differed significantly between groups in the pediatric sample (d = 0.68). In both samples, global MA increased classification accuracy compared to the model with clinical Conners' ADHD ratings alone. Regional diagnostic differences did not survive familywise correction for multiple comparisons. Global DTI metrics, particularly the mode of anisotropy, which is sensitive to crossing fibers, capture connectivity abnormalities in ADHD across both pediatric and adult samples. These findings highlight potential diffuse white matter microarchitecture differences in ADHD.
The Annals of Applied Statistics, 2015
Developmental Cognitive Neuroscience, 2015
To date, only one study has examined test-retest reliability of resting state fMRI (R-fMRI) in ch... more To date, only one study has examined test-retest reliability of resting state fMRI (R-fMRI) in children, none in clinical developing groups. Here, we assessed short-term test-retest reliability in a sample of 46 children (11-17.9 years) with attention-deficit/hyperactivity disorder (ADHD) and 57 typically developing children (TDC). Our primary test-retest reliability measure was the intraclass correlation coefficient (ICC), quantified for a range of R-fMRI metrics. We aimed to (1) survey reliability within and across diagnostic groups, and (2) compare voxel-wise ICC between groups. We found moderate-to-high ICC across all children and within groups, with higher-order functional networks showing greater ICC. Nearly all R-fMRI metrics exhibited significantly higher ICC in TDC than in children with ADHD for one or more regions. In particular, posterior cingulate and ventral precuneus exhibited group differences in ICC across multiple measures. In the context of overall moderate-to-high test-retest reliability in children, regional differences in ICC related to diagnostic groups likely reflect the underlying pathophysiology for ADHD. Our currently limited understanding of the factors contributing to inter- and intra-subject variability in ADHD underscores the need for large initiatives aimed at examining their impact on test-retest reliability in both clinical and developing populations.
Journal of Computational and Graphical Statistics a Joint Publication of American Statistical Association Institute of Mathematical Statistics Interface Foundation of North America, Jul 1, 2012
In linear regression with functional predictors and scalar responses, it may be advantageous, par... more In linear regression with functional predictors and scalar responses, it may be advantageous, particularly if the function is thought to contain features at many scales, to restrict the coefficient function to the span of a wavelet basis, thereby converting the problem into one of variable selection. If the coefficient function is sparsely represented in the wavelet domain, we may employ the well-known LASSO to select a relatively small number of nonzero wavelet coefficients. This is a natural approach to take but to date, the properties of such an estimator have not been studied. In this paper we describe the wavelet-based LASSO approach to regressing scalars on functions and investigate both its asymptotic convergence and its finite-sample performance through both simulation and real-data application. We compare the performance of this approach with existing methods and find that the wavelet-based LASSO performs relatively well, particularly when the true coefficient function is spiky. Source code to implement the method and data sets used in the study are provided as supplemental materials available online.
Journal of Computational and Graphical Statistics, Feb 12, 2014
Http Dx Doi Org 10 1198 016214507000000527, 2012
ABSTRACT
Journal of the American Statistical Association, Feb 1, 2006
Biostatistics (Oxford, England), Jan 29, 2016
SummaryMany modern neuroimaging studies acquire large spatial images of the brain observed sequen... more SummaryMany modern neuroimaging studies acquire large spatial images of the brain observed sequentially over time. Such data are often stored in the forms of matrices. To model these matrix-variate data we introduce a class of separable processes using explicit latent process modeling. To account for the size and two-way structure of the data, we extend principal component analysis to achieve dimensionality reduction at the individual level. We introduce necessary identifiability conditions for each model and develop scalable estimation procedures. The method is motivated by and applied to a functional magnetic resonance imaging study designed to analyze the relationship between pain and brain activity.
Journal of child psychology and psychiatry, and allied disciplines, Jan 22, 2016
Social anxiety disorder (SAD) typically onsets in adolescence and is associated with multiple imp... more Social anxiety disorder (SAD) typically onsets in adolescence and is associated with multiple impairments. Despite promising clinical interventions, most socially anxious adolescents remain untreated. To address this clinical neglect, we developed a school-based, 12-week group intervention for youth with SAD, Skills for Academic and Social Success (SASS). When implemented by psychologists, SASS has been found effective. To promote dissemination and optimize treatment access, we tested whether school counselors could be effective treatment providers. We randomized 138, ninth through 11th graders with SAD to one of three conditions: (a) SASS delivered by school counselors (C-SASS), (b) SASS delivered by psychologists (P-SASS), or (c) a control condition, Skills for Life (SFL), a nonspecific counseling program. Blind, independent, evaluations were conducted with parents and adolescents at baseline, post-intervention, and 5 months beyond treatment completion. We hypothesized that C-SASS...
Modern data-rich analyses may call for fitting a large number of nonparametric quantile regressio... more Modern data-rich analyses may call for fitting a large number of nonparametric quantile regressions. For example, growth charts may be constructed for each of a collection of variables, to identify those for which individuals with a disorder tend to fall in the tails of their age-specific distribution; such variables might serve as developmental biomarkers. When such a large set of analyses are carried out by penalized spline smoothing, reliable automatic selection of the smoothing parameter is particularly important. We show that two popular methods for smoothness selection may tend to overfit when estimating extreme quantiles as a smooth function of a predictor such as age; and that improved results can be obtained by multifold cross-validation or by a novel likelihood approach. A simulation study, and an application to a functional magnetic resonance imaging data set, demonstrate the favorable performance of our methods.
Journal of the American Academy of Child & Adolescent Psychiatry, 2015
Diffusion tensor imaging (DTI) can identify structural connectivity alterations in attention-defi... more Diffusion tensor imaging (DTI) can identify structural connectivity alterations in attention-deficit/hyperactivity disorder (ADHD). Most ADHD DTI studies have concentrated on regional differences in fractional anisotropy (FA) despite its limited sensitivity to complex white matter architecture and increasing evidence of global brain differences in ADHD. Here, we examine multiple DTI metrics in separate samples of children and adults with and without ADHD with a principal focus on global between-group differences. Two samples: adults with ADHD (n = 42) and without (n = 65) and children with ADHD (n = 82) and without (n = 80) were separately group matched for age, sex, and head motion. Five DTI metrics (FA, axial diffusivity, radial diffusivity, mean diffusivity, and mode of anisotropy) were analyzed via tract-based spatial statistics. Group analyses tested for diagnostic differences at the global (averaged across the entire white matter skeleton) and regional level for each metric. Robust global group differences in diffusion indices were found in adults, with the largest effect size for mode of anisotropy (MA; Cohen's d = 1.45). Global MA also differed significantly between groups in the pediatric sample (d = 0.68). In both samples, global MA increased classification accuracy compared to the model with clinical Conners' ADHD ratings alone. Regional diagnostic differences did not survive familywise correction for multiple comparisons. Global DTI metrics, particularly the mode of anisotropy, which is sensitive to crossing fibers, capture connectivity abnormalities in ADHD across both pediatric and adult samples. These findings highlight potential diffuse white matter microarchitecture differences in ADHD.
The Annals of Applied Statistics, 2015
Developmental Cognitive Neuroscience, 2015
To date, only one study has examined test-retest reliability of resting state fMRI (R-fMRI) in ch... more To date, only one study has examined test-retest reliability of resting state fMRI (R-fMRI) in children, none in clinical developing groups. Here, we assessed short-term test-retest reliability in a sample of 46 children (11-17.9 years) with attention-deficit/hyperactivity disorder (ADHD) and 57 typically developing children (TDC). Our primary test-retest reliability measure was the intraclass correlation coefficient (ICC), quantified for a range of R-fMRI metrics. We aimed to (1) survey reliability within and across diagnostic groups, and (2) compare voxel-wise ICC between groups. We found moderate-to-high ICC across all children and within groups, with higher-order functional networks showing greater ICC. Nearly all R-fMRI metrics exhibited significantly higher ICC in TDC than in children with ADHD for one or more regions. In particular, posterior cingulate and ventral precuneus exhibited group differences in ICC across multiple measures. In the context of overall moderate-to-high test-retest reliability in children, regional differences in ICC related to diagnostic groups likely reflect the underlying pathophysiology for ADHD. Our currently limited understanding of the factors contributing to inter- and intra-subject variability in ADHD underscores the need for large initiatives aimed at examining their impact on test-retest reliability in both clinical and developing populations.