 |
BIST 0535
Biostatistical Computing (3)
This course provides students with the ability to conduct statistical analysis using statistical software SAS and R. The primary subject is using the SAS programming language to solve a variety of database and statistical problems, and using R language to perform
statistical analysis. This is a highly applied course and students are expected to complete computer exercises each week.
|
BIST 0551
Applied Regression Analysis for Public Health Studies (3)
This course introduces students in graduate programs in public health to
regression analyses methods. The primary topics are simple, multiple linear regression models,
including analysis of covariance (ANCOVA), model diagnostics and model building. Logistic
regression for binary outcome will be introduced. The emphasis will be interpretation and
applications. Students will learn how to use SAS for implementing regression analyses.
Prerequisites: PHCO 0504 (Minimum Grade of B) and BIST 0535 (Minimum Grade of B); or BIST 0625 (Minimum Grade of B)
|
BIST 0610
Advanced Regression Methods for Public Health Studies (3)
This is an intermediate to advanced level course of regression
methods that emphasizes the theoretical concepts and applications of regression models for
public health studies. It is taught at BIST MPH/MS level. It covers simple and multiple linear
regression models, including analysis of variance (ANOVA) and co-variance (ANCOVA) and
binary regression logistic regression. Model building, model diagnostics, building hypothesis
testing, and interpretation as well as theoretical properties of parameter estimation and
inference will be taught. The theory part will use matrix and linear algebra.
Prerequisites: BIST 0613 and BIST 0535; or BIST 0613 and BIST 0625
|
BIST 0613
Biostatistics Theory I (3)
Statistics is a branch of science dealing with the collection, analysis,
interpretation, and presentation of numerical data. This course is an introduction to probability
modeling as a basis for statistical inference. The underlying strategy in prediction and inference
involves the determination of a probability modeling. This can be approached from two
perspectives: analytical (mathematical) and computational (algorithmic). It is important for
students to master both perspectives. Advanced calculus is sufficient background for the
analytical. This course serves to lay a foundation in statistical theory for M.S., M.PH, and
Ph.D. students so that they can pursue more advanced technical materials.
|
BIST 0614
Biostatistics Theory II (3)
This course will cover theory of estimation and hypothesis testing. Topics
include sampling distributions, sufficiency, unbiasedness, maximum likelihood methods
(estimation and tests). Emphasis is on the fundamental concepts of underlying theory.
Prerequisite: BIST 0613 (Minimum Grade of B).
|
BIST 0615
Applied Categorical Data Analysis (3)
Public health studies, especially those involving questionnaires, contain
large amounts of categorical data. This class provides an introduction to descriptive and
inferential statistics for univariate and multivariate categorical data with applications to
epidemiological and clinical studies. For 2 and 3-way contingency tables, measures of
association and tests for homogeneity between populations and independence of variables are
presented. Related tests of trend for ordinal data are studied. Loglinear and logistic regression
analyses are investigated for data sets with both nominal and ordinal variables.
Prerequisites: PHCO 0504 and BIST 0535; or BIST 0625
|
BIST 0625
Fundamentals of Biostatistics (3)
At the conclusion of this course, students will be able to distinguish
among the basic types of data; describe the normal curve and its major characteristics in
relation to parametric statistics; calculate descriptive statistics such as the mean, median,
variance, and standard error; describe the relationship of statistics to hypothesis testing;
understand the nature of Type I and Type II errors; explain the concept of statistical power and
how it can be calculated; apply basic statistical test procedures including t-tests, chi-square,
non-parametric tests, and correlation; decide which parametric or non-parametric test to apply
to test a statistical hypothesis; understand the concepts and applications of linear regression;
apply statistical software programs to solve common public health problems; and critically
review and comprehend basic statistical discussions in the public health literature. .
|
BIST 0627
Applied Survival Data Analysis (3)
This is a course in survival analysis that emphasizes concepts and
applications used in public health studies. The product limit estimator, the Cox proportional
hazard model, and parametric models will be discussed. Censoring and truncation patterns will
also be studied. Model building and checking will be discussed throughout.
Prerequisites: BIST 0551 (Minimum Grade of B) or BIST 0610 (Minimum Grade of B).
|
BIST 0630
Sampling Methods (3)
Data collection utilizing stratification, cluster sampling, multistage sampling, systematic sampling, or sampling with unequal probabilities of selection may be preferred to simple random sampling. Sampling strategies often introduce bias into traditional estimates of standard errors for these statistics. This course provides an introduction to theory and applications of methods for analysis that account for sampling design.
Prerequisites: PHCO 0504 and an intermediate statistics course.
|
BIST 0650
Applied Longitudinal Data Analysis (3)
Longitudinal data consists of multiple measures over time on a sample of
individuals. The analysis of longitudinal data requires much more sophisticated methodologies
due to the correlation introduced by repeated measurements. This course covers modern
statistical techniques for longitudinal data from an applied perspective. Emphasis will be on data
analysis and interpretation. Topics include characteristics of the longitudinal design, graphical
exploration of the mean and correlation structure, linear mixed effects models and multilevel
modelling, maximum likelihood and restricted maximum likelihood estimation, modeling the
variance-covariance structures, inference for random effects, logistic and Poisson mixed effects
model for binary and count data, marginal models and generalized estimating equations (GEE),
causal inference and time-dependent confounding (g-methods). Analysis of real and substantial
data sets using statistical software R (primarily) and SAS (secondarily) will be integrated
throughout.
Prerequisites: BIST 0551 (Minimum Grade of B) or BIST 0610 (Minimum Grade of B).
|
BIST 0655
Biocomputing II (3)
This extension of BIST 0654 provides students with more advanced computing-intensive methods for their future research needs. Topics include random variable generation, jackknife and bootstrap, Monte Carlo integration, Markov chains, Monte Carlo optimization, Metropolis-Hastings algorithm, and the Gibbs Sampler. SAS, R/Splus, BUGS (Bayesian inference using Gibbs sampling), together with Fortran, C/C++ will be used as programming tools.
Prerequisites: BIST 0535 and intermediate-level graduate courses in statistical inference and probability theory (e.g., BIST 0613 and BIST 0614). One year of calculus and some linear algebra. Permission from instructor.
|
BIST 0660
Clinical Trials: Design and Analysis of Medical Experiments (3)
This course will teach students essential concepts and application of a wide range of statistical
methods used in design, conduct, monitoring, and analysis of medical experiments.
Prerequisites are PHCO 0504 (Introduction to Biostatistics), BIST 0535 (Biometrics Computing)
BIST 0610 (Advanced Regression Methods for Public Health Studies), or their equivalents.
Students are expected to participate in group discussions during the class.
Prerequisites: BIST 0551 and BIST 0535; or BIST 0610 and BIST 0625.
|
BIST 0663
Statistical Learning in Biomedical Studies (3)
his is an intermediate level course of Statistical Learning. It emphasizes
the theoretical concepts and applications of statistical learning methods for biomedical studies.
It is taught at BIST MS level and also suitable for BIST doctoral students. It covers supervised
and unsupervised learning methods, including Ridge regression, LASSO, classification, decision
tree, random forest, SVM, neural network, deep learning, PCA and clustering models. Model
building, model assessment and selection, and interpretation as well as theoretical properties of
some models will be taught. R will be used for data analysis.
Prerequisites: BIST 0610 and BIST 0614.
|
BIST 0688
Statistical Methods in Genetics (3)
This is a course to acquaint students in public health with statistical methods for the analysis of genetic data arising in epidemiology, clinical trials, and laboratory experiments. Topics covered include genetic epidemiology, pedigree data, linkage and association analysis, phylogenetics, and the analysis of micro array
data.
Prerequisites: BIST 0535 or BIST 0625 and BIST 0610; or permission from instructor.
|
BIST 0690
Advanced Topics in Biostatistics (3)
An advanced-level course to provide students who are in research for doctoral thesis topics with
in-depth survey and synthesis of recent developments in biostatistics. Topics will be based on statistical research and applications of faculty of the department, which may include missing data analysis, sequential design and analysis, Bayesian statistics, bioinformatics analysis, mixed models, etc.
Prerequisite: BIST 0700 or permission from instructor.
|
BIST 0700
Advanced Theory of Biostatistics I (3)
This course extends probability and statistical theory covered in BIOS
0613 and 0614 to an advanced level. The topics include common stochastic processes in
biostatistics, likelihood construction, maximum likelihood estimation (MLE) and information
matrix, likelihood-based tests (Wald, Score and LRT), partial, conditional and marginal
likelihoods, Newton-Raphson algorithm, EM algorithm and extensions, convergence and limit
theorems, and asymptotic theory for likelihood-based methods. This course also includes the
inference for regression models and the theory of statistical estimation, unbiasedness,
consistency, and efficiency.
Prerequisites: An intermediate-level graduate course in statistical inference and probability theory (e.g., BIST 0613 and BIST 0614). One year of calculus and some linear algebra.
|
BIST 0701
Advanced Theory of Biostatistics II (3)
This course extends probability and statistical theory covered in BIST
0613 and 0614 to an advanced level, and continues the advanced inference topics arising in
Biostatistics presented in Advance Theory of Biostatistics I (BIST 700). Topics include
estimating equations, generalized estimating equation (GEE), restricted maximum likelihood
(REML) methods, jackknife and bootstrap methods, permutation and rank tests, and Bayesian
data analysis.
Prerequisites: An intermediate-level graduate course in statistical inference and probability theory (e.g., BIST 0613 and BIST 0614) or Advanced Theory of Biostatistics I (BIST 0700), or equivalent. One year of calculus and some linear algebra.
|
BIST 0714
Intermediate Biostatistics
his course is designed for students to take a deeper dive into modern
topics in biostatistics. The course is organized by aims of the research, rather than by statistical
methods. Topics include methods for randomized trials, observational studies, quasi-
experimental studies, and dealing with complexities such as missing data. Within each research
topic, a variety of methods will be learned, with an emphasis on assumptions, interpretation, and
strengths/weaknesses of each method. Through a combination of lectures and hands-on
assignments, students will learn to identify an appropriate study design to answer a given
research question, identify possible sources of bias, conduct an analysis appropriate to a given
study design and interpret results.
Doctoral student standing.
|
BIST 0720
Advanced Biostatistical Computing (3)
Statistical computing is an important part of Statistics/Biostatistics
Research. This course will cover advanced statistical computing techniques widely used in
Statistics/Biostatistics Research. Topics in this course will include dimensionality reduction,
random variable generation, Monte Carlo Integration, Markov chains, Monte Carlo Optimization,
EM algorithm, Metropolis-Hastings algorithm, the Gibbs sampler. R and/or C++ will be used as
programming tools in the class.
Prerequisites: BIST 0613, concurrent BIST 0614, and BIST 0610.
|
BIST 0721
Advanced Bayesian Inference (3)
This course focuses on modern Bayesian methods for analyzing biomedical and health-related data. The course begins with a brief review of Bayesian principles, conjugate and nonconjugate
priors, and computational algorithms (Gibbs; Metropolis Hastings; slice sampling) and emphasizes Bayesian nonparametric (BNP) methods. This includes popular priors over distributions (Dirichlet process, dependent Dirichlet process, and variations) and priors over functions (BART; Gaussian processes).
Doctoral student standing.
Prerequisites: BIST 0610 and BIST 0700.
|
BIST 0725
Generalized Linear Models (3)
This course covers the theory and inferential methods for generalized
linear models. Specific topics include linear, logistic, and Poisson regression; overdispersion;
likelihood-based inference and estimating equations; algorithms; and applications.
Prerequisites: BIST 0613, BIST 0614, and BIST 0610.
|
BIST 0727
Survival Analysis (3)
This is a course in survival analysis that emphasizes concepts and
applications used in public health studies. The product limit estimator, the Cox proportional
hazard model, and parametric models will be discussed. Censoring and truncation patterns will
also be studied. Model building and checking will be discussed throughout. This course is a
mathematically more advanced version of BIST 0627, with additional work required of students.
Prerequisites: BIST 0610 (Minimum Grade of B) and BIST 0614 (Minimum Grade of B).
|
BIST 0750
Longitudinal Data Analysis (3)
This course covers modern statistical techniques for longitudinal data analysis from a theoretical perspective. Emphasis will be on the theory of longitudinal data analysis and modeling. Topics include general and generalized linear models for longitudinal data, with the focus on the marginal, random effects and transition models, as well as missing data analysis.
Prerequisites: BIST 0610 (Minimum Grade of B) and BIST 0614 (Minimum Grade of B).
|
BIST 9619
Design of Experiments (3)
Teaches the fundamental principles of experimental design; completely randomized variance
component designs; randomized blocks; Latin squares; incomplete
blocks; partially hierarchic mixed-model experiments; factorial
experiments; fractional factorials; and response surface exploration.
Prerequisite: 01:960:484 or 401 or equivalent.
Cross-listed with 16:960:590. This course is offered through Rutgers School of Graduate Studies in New Brunswick.
|
BIST 9654
Applied Stochastic Processes (3)
Markov chains, recurrence, random walk, gambler's ruin, ergodic theorem and stationary distribution,
continuous time Markov chains, queuing problems, renewal processes, martingales, Markov processes,
Brownian motion, concepts in stochastic calculus, Ito's formula.
Prerequisites: Advanced calculus, and 16:960:580 or 582 or 592.
Cross-listed with 16:960:554. This course is offered through Rutgers School of Graduate Studies in New Brunswick.
|
BIST 9655
Nonparametric Statistics (3)
Introduction
and survey of distribution-free approaches to statistical inference.
Fisher's method of randomization; distribution-free test procedures
for means, variances, correlations, and trends; rank tests; relative
efficiency, asymptotic relative efficiency, and normal-score
procedures; binomial, hypergeometric distributions, and combinatorial
run theory. Also, tests of goodness-of-fit, including the
Kolmogorov-Smirnov and chi-square tests, contingency-table analysis,
tolerance sets, and Tchebycheffe-type inequalities. Emphasis on
applications.
Prerequsite: BIST 0613.
Cross-listed with 16:960:555. This course is offered through Rutgers School of Graduate Studies in New Brunswick.
|
BIST 9665
Applied Time-Series Analysis (3)
Model-based forecasting methods, autoregressive and moving average models, ARIMA, ARMAX, ARCH,
state-space models, estimation, forecasting and model validation, missing data, irregularly spaced time series, parametric and nonparametric bootstrap methods for time series, multiresolution analysis of spatial and time series signals, time-varying models and wavelets.
Prerequisite: Level V statistics or permission of instructor.
Cross-listed with 16:960:565. This course is offered through Rutgers School of Graduate Studies in New Brunswick.
|
BIST 9667
Applied Multivariate Analysis (3)
Methods dimension reductiony, including principal components, factor analysis, and multidimensional scaling; correlation techniques, including partial, multiple, and canonical correlation; classification and clustering methods. Emphasis on data analytic issues, concepts, and methods (e.g., graphical techniques) and on applications drawn from several areas, including behavioral, management, and physical and engineering
sciences.
Prerequisite: Level V statistics or permission of instructor.
Cross-listed with 16:960:567. This course is offered through Rutgers School of Graduate Studies in New Brunswick.
|
BIST 9686 and BIST 9687
Interpretation of Data I, II (3)
Use of various computer-based techniques, including graphical, to understand and interpret data sets.
Exposure to, and intuitive understanding of, some basic techniques for the analysis of multivariate,
categorical, and time-series data as well as other miscellaneous applications of statistical procedures.
Prerequisite: Level IV statistics.
Cross-listed with 16:960:586 and 16:960:587. This course is offered through Rutgers School of Graduate Studies in New Brunswick.
|
BIST 9688
Data Mining (3)
Databases and data warehousing, exploratory data analysis and visualization, an overview of data-mining algorithms, modeling for data mining, descriptive modeling, predictive modeling, pattern and rule discovery, text mining, Bayesian data mining, observational studies.
Prerequisite: 16:960:567 or 568 or permission of instructor.
Cross-listed with 16:960:588. This course is offered through Rutgers School of Graduate Studies in New Brunswick.
|
BIST 9691
Advanced Design of Experiments (3)
Strategy of experimentation, screening designs, factorial designs, response surface methodology, evolutionary operation, mixture designs, incomplete blocking designs, computer-aided experimental designs, and design optimality criteria.
Prerequisite: 16:960:590. Recommended: 16:960:563.
Cross-listed with 16:960:591. This course is offered through Rutgers School of Graduate Studies in New Brunswick.
|
|
