Hyokyoung Grace Hong

Quantile-adaptive model-free variable screening for high-dimensional heterogeneous data

We introduce a quantile-adaptive framework for nonlinear variable screening with high-dimensional heterogeneous data. This framework has two distinctive features: (1) it allows the set of active variables to vary across quantiles, thus making it more flexible to accommodate heterogeneity; (2) it is model-free and avoids the difficult task of specifying the form of a statistical model in a high dimensional space. Our nonlinear independence screening procedure employs spline approximations to model the marginal effects at a quantile level of interest. Under appropriate conditions on the quantile functions without requiring the existence of any moments, the new procedure is shown to enjoy the sure screening property in ultra-high dimensions. Furthermore, the quantile-adaptive framework can naturally handle censored data arising in survival analysis. We prove that the sure screening property remains valid when the response variable is subject to random right censoring. Numerical studies confirm the fine performance of the proposed method for various semiparametric models and its effectiveness to extract quantile-specific information from heteroscedastic data.

Personality and alcohol use: The role of impulsivity

Research has shown that personality traits associated with impulsivity influence alcohol use during emerging adulthood, yet relatively few studies have examined how distinct facets of impulsivity are associated with alcohol use and abuse. We examine the influence of impulsivity traits on four patterns of alcohol use including frequency of alcohol use, alcohol-related problems, binge drinking, and alcohol use disorders (AUDs) in a community sample of young individuals (N=190). In multivariate regression analyses that controlled for peer and parental alcohol use, psychological distress, and developmental correlates (i.e., college, marriage, employment) in emerging adulthood, we found that urgency and sensation seeking were consistently related to all four constructs of alcohol use. The present study suggests that distinct impulsivity traits may play different roles in escalation of alcohol use and development of AUDs during emerging adulthood.

Superficial x-ray in the treatment of basal and squamous cell carcinomas: A viable option in select patients

BACKGROUND Effective nonsurgical modalities are limited in the treatment of basal cell carcinoma (BCC) and squamous cell carcinoma (SCC). OBJECTIVE We sought to evaluate the efficacy and viability of superficial x-ray therapy in the treatment of BCC and SCC in an outpatient setting. METHODS A retrospective analysis was performed on 1715 histologically confirmed primary cutaneous BCC and SCC treated with superficial x-ray therapy at Dermatology Associates of Tallahassee in Florida between 2000 and 2010. RESULTS Of the 1715 tumors reviewed during this period, 712 were histologically proven BCC (631 nodular and 81 superficial), 994 were SCC (861 SCC in situ and 133 invasive SCC), and 9 displayed distinct features of both BCC and SCC in the same biopsy specimen. Kaplan-Meier estimates (with 95% confidence intervals) of cumulative recurrence rates of all tumors at 2 and 5 years were 1.9% (1%-2.7%) and 5.0% (3.2%-6.7%), respectively; of BCC at 2 and 5 years were 2% (0.8%-3.3%) and 4.2% (1.9%-6.4%), respectively; and of all SCC at 2 and 5 years were 1.8% (0.8%-2.8%) and 5.8% (2.9%-8.7%), respectively. Tumors on male patients and those with a diameter greater than 2 cm were associated with a statistically significant increase in recurrence likelihood. LIMITATIONS This study represents only patients treated in 1 dermatology office in North Florida and may not be representative of the general patient population. CONCLUSIONS Superficial x-ray therapy remains a viable nonsurgical option for the treatment of primary BCC and SCC in patients where surgical intervention is declined, unadvisable, or potentially associated with significant cosmetic or functional limitations.

Childhood sexual abuse and adolescent substance use: A latent class analysis

Children who have exposure to child sexual abuse (CSA) are at particular risk for developing substance abuse in adolescence, but the extent to which CSA may shape patterns of adolescent substance use remains uncertain. The aim of this paper is to characterize the variations in patterns of adolescent substance use and to examine the association between CSA and qualitatively distinct patterns of adolescent substance use. Latent class analyses identified homogenous groups of adolescents with similar patterns of substance use using a sample of 1019 adolescents (mean age: 15.9 years; range: 13-18) who were selected from five publicly funded service systems. Different patterns of latent class structures were identified in boys and girls (a 4-class solution for girls and a 3-class solution for boys). CSA was associated with an increased risk of being a heavy polysubstance user in girls, even after adjustment of age, race/ethnicity, parental substance use, sibling use, peer use, psychopathology and other forms of childhood maltreatment including physical abuse and neglect. Findings indicate that female victims of CSA who are involved with public service systems are at high risk for developing multiple-substance use in adolescence.

An Examination of Pathways from Childhood Maltreatment to Adolescent Binge Drinking

Considerable clinical and empirical evidence has accumulated over the past decades indicating that there is a strong association between childhood maltreatment and heavy episodic drinking in adolescence, but there is a paucity of empirically based knowledge about the processes linking the association. The aim of this paper is to examine mechanisms that might account for the association between childhood maltreatment and heavy episodic drinking in adolescence. Using a nationally representative sample of adolescents (ages ranging 12-21; N = 6,337), this study examined the role of individual self-regulatory processes in the associations, controlling for age, gender, race/ethnicity, peer substance use, parental alcoholism, and parent-child conflict. Factor analyses were used to test the measurement structure of self-regulatory processes. Findings confirmed the association between childhood maltreatment and heavy episodic drinking in adolescence. Structural modeling analyses indicated indirect effects for childhood maltreatment primarily through poor self-regulatory processes and peer substance use. Implications for future research are discussed.

Comorbidity of chronic diseases in the elderly

Joint modeling of multiple health related random variables is essential to develop an understanding for the public health consequences of an aging population. This is particularly true for patients suffering from multiple chronic diseases. The contribution is to introduce a novel model for multivariate data where some response variables are discrete and some are continuous. It is based on pair copula constructions (PCCs) and has two major advantages over existing methodology. First, expressing the joint dependence structure in terms of bivariate copulas leads to a computationally advantageous expression for the likelihood function. This makes maximum likelihood estimation feasible for large multidimensional data sets. Second, different and possibly asymmetric bivariate (conditional) marginal distributions are allowed which is necessary to accurately describe the limiting behavior of conditional distributions for mixed discrete and continuous responses. The advantages and the favorable predictive performance of the model are demonstrated using data from the Second Longitudinal Study of Aging (LSOA II).

Conditional screening for ultra-high dimensional covariates with survival outcomes

Identifying important biomarkers that are predictive for cancer patients’ prognosis is key in gaining better insights into the biological influences on the disease and has become a critical component of precision medicine. The emergence of large-scale biomedical survival studies, which typically involve excessive number of biomarkers, has brought high demand in designing efficient screening tools for selecting predictive biomarkers. The vast amount of biomarkers defies any existing variable selection methods via regularization. The recently developed variable screening methods, though powerful in many practical setting, fail to incorporate prior information on the importance of each biomarker and are less powerful in detecting marginally weak while jointly important signals. We propose a new conditional screening method for survival outcome data by computing the marginal contribution of each biomarker given priorily known biological information. This is based on the premise that some biomarkers are known to be associated with disease outcomes a priori. Our method possesses sure screening properties and a vanishing false selection rate. The utility of the proposal is further confirmed with extensive simulation studies and analysis of a diffuse large B-cell lymphoma dataset. We are pleased to dedicate this work to Jack Kalbfleisch, who has made instrumental contributions to the development of modern methods of analyzing survival data.

Bayesian Tobit quantile regression model for medical expenditure panel survey data

High expenditure on healthcare is an important segment of the U.S. economy, making healthcare cost modelling valuable in decision-making processes over a wide array of domains. In this paper, we analyze medical expenditure panel survey (MEPS) data. Tobit regression model has been popularly used for the medical expenditures. However, it is no longer sufficient for the MEPS data because: (i) the distribution of the expenditures shows skewness, heavy tails and heterogeneity; (ii) most predictors are categorical, including binary, nominal and ordinal variables; (iii) there are a few predictors which may be nonlinearly related to the response. We therefore propose a Bayesian Tobit quantile regression model to describe a complete distributional view on how the medical expenditures depend on the various predictors. Specifically, we assume an asymmetric Laplace error distribution to adapt the quantile regression to a Bayesian setting. Then, we propose a modified group Lasso for categorical factor selection, and a smoothing Gaussian prior for modelling the nonlinear effects. The estimates and their uncertainties are obtained using an efficient Monte Carlo Markov Chain sampling method. The effectiveness of our approach is demonstrated by modelling 2007 MEPS data.

A multi-index model for quantile regression with ordinal data

In this paper, we propose a quantile approach to the multi-index semiparametric model for an ordinal response variable. Permitting non-parametric transformation of the response, the proposed method achieves a root-n rate of convergence and has attractive robustness properties. Further, the proposed model allows additional indices to model the remaining correlations between covariates and the residuals from the single-index, considerably reducing the error variance and thus leading to more efficient prediction intervals (PIs). The utility of the model is demonstrated by estimating PIs for functional status of the elderly based on data from the second longitudinal study of aging. It is shown that the proposed multi-index model provides significantly narrower PIs than competing models. Our approach can be applied to other areas in which the distribution of future observations must be predicted from ordinal response data.

Efficient quantile marginal regression for longitudinal data with dropouts

In many biomedical studies independent variables may affect the conditional distribution of the response differently in the middle as opposed to the upper or lower tail. Quantile regression evaluates diverse covariate effects on the conditional distribution of the response with quantile-specific regression coefficients. In this paper, we develop an empirical likelihood inference procedure for longitudinal data that accommodates both the within-subject correlations and informative dropouts under missing at random mechanisms. We borrow the matrix expansion idea of the quadratic inference function and incorporate the within-subject correlations under an informative working correlation structure. The proposed procedure does not assume the exact knowledge of the true correlation structure nor does it estimate the parameters of the correlation structure. Theoretical results show that the resulting estimator is asymptotically normal and more efficient than one attained under a working independence correlation structure. We expand the proposed approach to account for informative dropouts under missing at random mechanisms. The methodology is illustrated by empirical studies and a real-life example of HIV data analysis.