Applied Statistics Commons^™

Deep Neural Network Architectures For Music Genre Classification, Kai Middlebrook, Shyam Sudhakaran, Kunal Sonar, David Guy Brizan

Creative Activity and Research Day - CARD

With the recent advancements in technology, many tasks in fields such as computer vision, natural language processing, and signal processing have been solved using deep learning architectures. In the audio domain, these architectures have been used to learn musical features of songs to predict: moods, genres, and instruments. In the case of genre classification, deep learning models were applied to popular datasets--which are explicitly chosen to represent their genres--and achieved state-of-the-art results. However, these results have not been reproduced on less refined datasets. To this end, we introduce an un-curated dataset which contains genre labels and 30-second audio previews for …

The Andersen Likelihood Ratio Test With A Random Split Criterion Lacks Power, Georg Krammer

Journal of Modern Applied Statistical Methods

The Andersen LRT uses sample characteristics as split criteria to evaluate Rasch model fit, or theory driven hypothesis testing for a test. The power and Type I error of a random split criterion was evaluated with a simulation study. Results consistently show a random split criterion lacks power.

Weighted Version Of Generalized Inverse Weibull Distribution, Sofi Mudiasir, S. P. Ahmad

Journal of Modern Applied Statistical Methods

Weighted distributions are used in many fields, such as medicine, ecology, and reliability. A weighted version of the generalized inverse Weibull distribution, known as weighted generalized inverse Weibull distribution (WGIWD), is proposed. Basic properties including mode, moments, moment generating function, skewness, kurtosis, and Shannon’s entropy are studied. The usefulness of the new model was demonstrated by applying it to a real-life data set. The WGIWD fits better than its submodels, such as length biased generalized inverse Weibull (LGIW), generalized inverse Weibull (GIW), inverse Weibull (IW) and inverse exponential (IE) distributions.

Calibration Of Measurements, Edward Kroc, Bruno D. Zumbo

Journal of Modern Applied Statistical Methods

Traditional notions of measurement error typically rely on a strong mean-zero assumption on the expectation of the errors conditional on an unobservable “true score” (classical measurement error) or on the data themselves (Berkson measurement error). Weakly calibrated measurements for an unobservable true quantity are defined based on a weaker mean-zero assumption, giving rise to a measurement model of differential error. Applications show it retains many attractive features of estimation and inference when performing a naive data analysis (i.e. when performing an analysis on the error-prone measurements themselves), and other interesting properties not present in the classical or Berkson cases. Applied …

Estimation Of Mean With Two-Parameter Ratio-Product-Ratio Estimator In Double Sampling Using Ancillary Information Under Non-Response, Surya K. Pal, Housila P. Singh

Journal of Modern Applied Statistical Methods

Ratio-product-ratio estimators with two parameters in double sampling under non-response are considered along with their properties. Practical conditions are obtained in which the suggested estimators are more proficient than other existing estimators. An example is given.

Capturing Heterogeneity Of Covariate Effects In Hidden Subpopulations In The Presence Of Censoring And Large Number Of Covariates, Farhad Shokoohi, Abbas Khalili, Masoud Asgharian, Shili Lin

Mathematical Sciences Faculty Research

The advent of modern technology has led to a surge of high-dimensional data in biology and health sciences such as genomics, epigenomics and medicine. The high-grade serous ovarian cancer (HGS-OvCa) data reported by The Cancer Genome Atlas (TCGA) Research Network is one example. The TCGA and other research groups have analyzed several aspects of these data. Here we study the relationship between Disease Free Time (DFT) after surgery among ovarian cancer patients and their DNA methylation profiles of genomic features. Such studies pose additional challenges beyond the typical big data problem due to population substructure and censoring. Despite the availability …

A More Powerful Unconditional Exact Test Of Homogeneity For 2 × C Contingency Table Analysis, Louis Ehwerhemuepha, Heng Sok, Cyril Rakovski

Mathematics, Physics, and Computer Science Faculty Articles and Research

The classical unconditional exact p-value test can be used to compare two multinomial distributions with small samples. This general hypothesis requires parameter estimation under the null which makes the test severely conservative. Similar property has been observed for Fisher's exact test with Barnard and Boschloo providing distinct adjustments that produce more powerful testing approaches. In this study, we develop a novel adjustment for the conservativeness of the unconditional multinomial exact p-value test that produces nominal type I error rate and increased power in comparison to all alternative approaches. We used a large simulation study to empirically estimate the …

Sensitivity Analyses For Tumor Growth Models, Ruchini Dilinika Mendis

Masters Theses & Specialist Projects

This study consists of the sensitivity analysis for two previously developed tumor growth models: Gompertz model and quotient model. The two models are considered in both continuous and discrete time. In continuous time, model parameters are estimated using least-square method, while in discrete time, the partial-sum method is used. Moreover, frequentist and Bayesian methods are used to construct confidence intervals and credible intervals for the model parameters. We apply the Markov Chain Monte Carlo (MCMC) techniques with the Random Walk Metropolis algorithm with Non-informative Prior and the Delayed Rejection Adoptive Metropolis (DRAM) algorithm to construct parameters' posterior distributions and then …

Daily And Seasonal Variability Of Offshore Wind Power On The Central California Coast And Statewide Demand, Matthew Douglas Kehrli

Physics

No abstract provided.

Efficient Class Of Estimators For Finite Population Mean Using Auxiliary Information In Two-Occasion Successive Sampling, G. N. Singh, Mohd Khalid

Journal of Modern Applied Statistical Methods

In the case of sampling on two occasions, a class of estimators is considered which uses information on the first occasion as well as the second occasion in order to estimate the population means on the current (second) occasion. The usefulness of auxiliary information in enhancing the efficiency of this estimation is examined through the class of proposed estimators. Some properties of the class of estimators and a strategy of optimum replacement are discussed. The proposed class of estimators were empirically compared with the sample mean estimator in the case of no matching. The established optimum estimator, which is a …

Latent Choice Models To Account For Misclassification Errors In Discrete Transportation Data, Lacramioara Elena Balan

Civil & Environmental Engineering Theses & Dissertations

One of the most fundamental tasks when it comes to analyzing data using statistical methods is to understand the relationship between the explanatory variables and the outcome. Misclassification of explanatory variables is a common risk when using statistical modeling techniques. In this dissertation, we define ‘misclassification,’ as a response that is reported or recorded in the wrong category; for example, a variable is registered as a one when it should have the value zero. Misclassification can easily happen in any data; for example, in an interview setting where the respondent misunderstands the question or the interviewer checks the wrong box. …

Spatio-Temporal Cluster Detection And Local Moran Statistics Of Point Processes, Jennifer L. Matthews

Mathematics & Statistics Theses & Dissertations

Moran's index is a statistic that measures spatial dependence, quantifying the degree of dispersion or clustering of point processes and events in some location/area. Recognizing that a single Moran's index may not give a sufficient summary of the spatial autocorrelation measure, a local indicator of spatial association (LISA) has gained popularity. Accordingly, we propose extending LISAs to time after partitioning the area and computing a Moran-type statistic for each subarea. Patterns between the local neighbors are unveiled that would not otherwise be apparent. We consider the measures of Moran statistics while incorporating a time factor under simulated multilevel Palm distribution, …

Assessment And Correction Of Lidar-Derived Dems In The Coastal Marshes Of Louisiana, William M. Lauve

LSU Master's Theses

The onset of airborne light detection and ranging (lidar) has resulted in expansive, precise digital elevation models (DEMs). DEMs are essential for modeling complex systems, such as the coastal land margin of Louisiana. They are used for many applications (e.g. tide, storm surge, and ecological modeling) and by diverse groups (e.g. state and federal agencies, NGOs, and academia). However, in a marsh environment, it is difficult for airborne lidar to produce accurate bare-earth measurements and even accurate elevations are rarely verified by ground truth data. The accuracy of lidar in marshes is limited by the sensor’s resolution …

Tobacco Smoking And Dementia In A Kentucky Cohort: A Competing Risk Analysis, Erin L. Abner, Peter T. Nelson, Gregory A. Jicha, Gregory E. Cooper, David W. Fardo, Frederick A. Schmitt, Richard J. Kryscio

Epidemiology and Environmental Health Faculty Publications

Tobacco smoking was examined as a risk for dementia and neuropathological burden in 531 initially cognitively normal older adults followed longitudinally at the University of Kentucky’s Alzheimer’s Disease Center. The cohort was followed for an average of 11.5 years; 111 (20.9%) participants were diagnosed with dementia, while 242 (45.6%) died without dementia. At baseline, 49 (9.2%) participants reported current smoking (median pack-years = 47.3) and 231 (43.5%) former smoking (median pack-years = 24.5). The hazard ratio (HR) for dementia for former smokers versus never smokers based on the Cox model was 1.64 (95% CI: 1.09, 2.46), while the HR for …

Jmasm 51: Bayesian Reliability Analysis Of Binomial Model – Application To Success/Failure Data, M. Tanwir Akhtar, Athar Ali Khan

Journal of Modern Applied Statistical Methods

Reliability data are generated in the form of success/failure. An attempt was made to model such type of data using binomial distribution in the Bayesian paradigm. For fitting the Bayesian model both analytic and simulation techniques are used. Laplace approximation was implemented for approximating posterior densities of the model parameters. Parallel simulation tools were implemented with an extensive use of R and JAGS. R and JAGS code are developed and provided. Real data sets are used for the purpose of illustration.

Comparative Clinical Outcomes Between Direct Oral Anticoagulants And Warfarin Among Elderly Patients With Non-Valvular Atrial Fibrillation In The Cms Medicare Population, Alpesh Amin, Oluwaseyi Dina, Allison Keshishian, Amol Dhamane, Anagha Nadkarni, Eric Carda, Cristina Russ, Lisa Rosenblatt, Jack Mardekian, Huseyin Yuce, Christine L. Baker

Publications and Research

Atrial fibrillation (AF) prevalence increases with age; > 80% of US adults with AF are aged ≥ 65 years. Compare the risk of stroke/systemic embolism (SE), major bleeding (MB), net clinical outcome (NCO), and major adverse cardiac events (MACE) among elderly non-valvular AF (NVAF) Medicare patients prescribed direct oral anticoagulants (DOACs) vs warfarin. NVAF patients aged ≥ 65 years who initiated DOACs (apixaban, dabigatran, and rivaroxaban) or warfarin were selected from 01JAN2013-31DEC2015 in CMS Medicare data. Propensity score matching was used to balance DOAC and warfarin cohorts. Cox proportional hazards models estimated the risk of stroke/SE, MB, NCO, and MACE. 37,525 …

Six-Month Outcome Of Transient Ischemic Attack And Its Mimics, Alireza Sadighi, Vida Abedi, Alia C. Stanciu, Nada El Andary, Mihai Banciu, Neil Holland, Ramin Zand

Faculty Journal Articles

Background and Objective: Although the risk of recurrent cerebral ischemia is higher after a transient ischemic attack (TIA), there is limited data on the outcome of TIA mimics. The goal of this study is to compare the 6-month outcome of patients with negative and positive diffusion-weighted imaging (DWI) TIAs (DWI-neg TIA vs. DWI-pos TIA) and also TIA mimics.

Methods: We prospectively studied consecutive patients with an initial diagnosis of TIA in our tertiary stroke centers in a 2-year period. Every included patient had an initial magnetic resonance (MR) with DWI and one-, three-, and six-month follow-up visits. The primary outcome …

A Random Forests Approach To Assess Determinants Of Central Bank Independence, Maddalena Cavicchioli, Angeliki Papana, Ariadni Papana Dagiasis, Barbara Pistoresi

Journal of Modern Applied Statistical Methods

A non-parametric efficient statistical method, Random Forests, is implemented for the selection of the determinants of Central Bank Independence (CBI) among a large database of economic, political, and institutional variables for OECD countries. It permits ranking all the determinants based on their importance in respect to the CBI and does not impose a priori assumptions on potential nonlinear relationships in the data. Collinearity issues are resolved, because correlated variables can be simultaneously considered.

Maximum Likelihood Estimation For The Generalized Pareto Distribution And Goodness-Of-Fit Test With Censored Data, Minh H. Pham, Chris Tsokos, Bong-Jin Choi

Journal of Modern Applied Statistical Methods

The generalized Pareto distribution (GPD) is a flexible parametric model commonly used in financial modeling. Maximum likelihood estimation (MLE) of the GPD was proposed by Grimshaw (1993). Maximum likelihood estimation of the GPD for censored data is developed, and a goodness-of-fit test is constructed to verify an MLE algorithm in R and to support the model-validation step. The algorithms were composed in R. Grimshaw’s algorithm outperforms functions available in the R package ‘gPdtest’. A simulation study showed the MLE method for censored data and the goodness-of-fit test are both reliable.

Unified Methods For Feature Selection In Large-Scale Genomic Studies With Censored Survival Outcomes, Lauren Spirko-Burns, Karthik Devarajan

COBRA Preprint Series

One of the major goals in large-scale genomic studies is to identify genes with a prognostic impact on time-to-event outcomes which provide insight into the disease's process. With rapid developments in high-throughput genomic technologies in the past two decades, the scientific community is able to monitor the expression levels of tens of thousands of genes and proteins resulting in enormous data sets where the number of genomic features is far greater than the number of subjects. Methods based on univariate Cox regression are often used to select genomic features related to survival outcome; however, the Cox model assumes proportional hazards …

A Simulation Study Of Diagnostics For Bias In Non-Probability Samples, Philip S. Boonstra, Roderick Ja Little, Brady T. West, Rebecca R. Andridge, Fernanda Alvarado-Leiton

The University of Michigan Department of Biostatistics Working Paper Series

A non-probability sampling mechanism is likely to bias estimates of parameters with respect to a target population of interest. This bias poses a unique challenge when selection is 'non-ignorable', i.e. dependent upon the unobserved outcome of interest, since it is then undetectable and thus cannot be ameliorated. We extend a simulation study by Nishimura et al. [International Statistical Review, 84, 43--62 (2016)], adding a recently published statistic, the so-called 'standardized measure of unadjusted bias', which explicitly quantifies the extent of bias under the assumption that a specified amount of non-ignorable selection exists. Our findings suggest that this new …

Bayesian Approximation Techniques For Scale Parameter Of Laplace Distribution, Uzma Jan, S. P. Ahmad

Journal of Modern Applied Statistical Methods

The Bayesian estimation of the scale parameter of a Laplace Distribution is obtained using two approximation techniques, like Normal approximation and Tierney and Kadane (T-K) approximation, under different informative priors.

Can One Test Fit All? Responses To The Article “Striving For Simple But Effective Advice For Comparing The Central Tendency Of Two Populations” (Ruxton & Neuhäuser, 2018), Diep Nguyen, Eun Sook Kim, Yi-Hsin Chen

Journal of Modern Applied Statistical Methods

Responses to suggestions made by Ruxton & Neuhäuser (2018) regarding Nguyen et al. (2016) are given.

Associations Of Hemodynamics, Morphology, And Patient Characteristics With Aneurysm Rupture Stratified By Aneurysm Location, Felicitas J. Detmer, Bong Jae Chung, Carlos Jimenez, Farid Hamzei-Sichani, David Kallmes, Christopher Putman, Juan R. Cebral

Department of Applied Mathematics and Statistics Faculty Scholarship and Creative Works

Purpose: The mechanisms of cerebral aneurysm rupture are not fully understood. We analyzed the associations of hemodynamics, morphology, and patient age and gender with aneurysm rupture stratifying by location. Methods: Using image-based models, 20 hemodynamic and 17 morphological parameters were compared in 1931 ruptured and unruptured aneurysms with univariate logistic regression. Rupture rates were compared between males and females as well as younger and older patients and bifurcation versus sidewall aneurysms for different aneurysm locations. Subsequently, associations between hemodynamics and morphology and patient as well as aneurysm characteristics were analyzed for aneurysms at five locations. Results: Compared to unruptured aneurysms, …

On The Conditional And Unconditional Type I Error Rates And Power Of Tests In Linear Models With Heteroscedastic Errors, Patrick J. Rosopa, Alice M. Brawley, Theresa P. Atkinson, Stephen A. Robertson

Journal of Modern Applied Statistical Methods

Preliminary tests for homoscedasticity may be unnecessary in general linear models. Based on Monte Carlo simulations, results suggest that when testing for differences between independent slopes, the unconditional use of weighted least squares regression and HC4 regression performed the best across a wide range of conditions.

Φ-Divergence Loss-Based Artificial Neural Network, R. L. Salamwade, D. M. Sakate, S. K. Mathur

Journal of Modern Applied Statistical Methods

Artificial Neural Networks (ANNs) can fit non-linear functions and recognize patterns better than several standard techniques. Performance of ANNs is measured by using loss functions. Phi-divergence estimator is generalization of maximum likelihood estimator and it possesses all its properties. A neural network is proposed which is trained using phi-divergence loss.

Robust Ancova, Curvature, And The Curse Of Dimensionality, Rand Wilcox

Journal of Modern Applied Statistical Methods

There is a substantial collection of robust analysis of covariance (ANCOVA) methods that effectively deals with non-normality, unequal population slope parameters, outliers, and heteroscedasticity. Some are based on the usual linear model and others are based on smoothers (nonparametric regression estimators). However, extant results are limited to one or two covariates. A minor goal here is to extend a recently-proposed method, based on the usual linear model, to situations where there are up to six covariates. The usual linear model might provide a poor approximation of the true regression surface. The main goal is to suggest a method, based on …

A Strategy For Using Bias And Rmse As Outcomes In Monte Carlo Studies In Statistics, Michael Harwell

Journal of Modern Applied Statistical Methods

To help ensure important patterns of bias and accuracy are detected in Monte Carlo studies in statistics this paper proposes conditioning bias and root mean square error (RMSE) measures on estimated Type I and Type II error rates. A small Monte Carlo study is used to illustrate this argument.

A Robust Nonparametric Measure Of Effect Size Based On An Analog Of Cohen's D, Plus Inferences About The Median Of The Typical Difference, Rand Wilcox

Journal of Modern Applied Statistical Methods

The paper describes a nonparametric analog of Cohen's d, Q. It is established that a confidence interval for Q can be computed via a method for computing a confidence interval for the median of D = X₁ − X₂, which in turn is related to making inferences about P(X₁ < X₂).

Should We Give Up On Causality?, Tom Knapp

Journal of Modern Applied Statistical Methods

No abstract provided.