Categorical Data Analysis

Do you work with binary, categorical, or count data?  You'd be a rare analyst if you didn't.  There are lots of details to consider - when to do exact tests, what they are, how to handle ordinal data, the role of modeling as opposed to significance testing, what to do when the assumptions required for Poisson regression fail, and much more.  Learn more in online course on "Categorical Data Analysis" (Brian Marx) at Statistis.com.

"Categorical Data Analysis," covers the analysis of contingency table data (tabular data in which the cell entries represent counts of subjects or items falling into certain categories). Topics include tests for independence (comparing proportions as well as chi-square), exact methods, treatment of ordered data, as well as models: Poisson regression for count responses and logistic/probit regression for binomial responses. Both 2-way and 3-way tables are covered. The focus will be on interpretation of models rather than the theory behind them.  Exercises involving the use of standard statistical software are part of the course; no particular software program is required. For more details please visit at http://www.statistics.com/categorical1/.

Who Should Take This Course:

Anyone who needs to analyze data in which the response is in categorical form. Market researchers, medical researchers, surveyors, those who study education assessment data, quality control specialists, life scientists, environmental scientists, ecologists.

Course Program:

Course outline: The course is structured as follows

SESSION 1: Overview of Categorical Responses

• Binomial and multinomial distributions
• Maximum Likelihood
• Test of proportions

SESSION 2: Structure of Contingency Tables

• Joint, marginal and conditional probabilities
• Odds ratio and relative risk
• Test of independence
• Ordinal data
• Exact testing
• Three-way tables
• Conditional independence and homogenous association

SESSION 3: Generalized Linear Models

• Binary data: logistic and probit models
• Poisson regression for count data
• Model checking

SESSION 4: Applications and Interpretations for Logistic Regression

• Inference
• Categorical predictors
• Multiple logistic regression

Dr. Brian Marx is Professor of Statistics at Louisiana State University, and has taught Categorical Data Analysis for over ten years. He is currently serving as Chair of the Statistical Modeling Society and is the Coordinating Editor of "Statistical Modeling:  An International Journal."

You will be able to ask questions and exchange comments with Dr. Brian Marx via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune.

For India Registration and pricing, please visit us at www.india.statistics.com.

Email: info@c-elt.com

Call: 020 66009116

Websites:

www.india.statistics.com

www.c-elt.com

Survival Analysis

Do telephone poles need to be treated with creosote every 2 years?  Or will every 5 years do? Do heart patients do better with plain stents, or stents coated with medication?  We typically measure the outcomes of treatments like this in terms of "survival times" (in medicine) or "time to event" (more broadly).  Our online course in the subject is called "Survival Analysis," by Dr. David Kleinbaum and Prof. Matthew Strickland.

"Survival Analysis" describes the various methods used for modeling and evaluating survival data, also called time-to-event data.  Survival models are used in a variety of health and social sciences, including biostatistics, epidemiology, anthropology, sociology, psychology and economics.  In engineering applications, the topic is called "time-to-failure" analysis.  General statistical concepts and methods discussed in this course include survival and hazard functions, Kaplan-Meier graphs, log-rank and related tests, Cox proportional hazards model, and the extended Cox model for time-varying covariates. For more details please visit at http://www.statistics.com/survival/.

Who Should Take This Course:

Investigators designing, conducting or analyzing medical studies or clinical trials. Researchers in any field (including engineering) working with data on how long things last.

Course Program:

Course outline: The course is structured as follows

SESSION 1:

• An overview of survival analysis methods
• Censoring
• Key terms: survival and hazard functions
• Goals of a survival analysis
• Data layout for the computer
• Data layout for theory
• Descriptive statistics for survival analysis- the hazard ratio
• Graphing survival data- Kaplan Meier
• The Log Rank and related tests.

SESSION 2:

• Introduction to the Cox Proportional Hazards (PH) model- computer example
• Model definition and features
• Maximum likelihood estimation for the Cox PH model
• Computing the hazard ratio in the Cox PH model
• The PH assumption
• Adjusted survival curves
• Checking the proportional hazard assumption
• The likelihood function for the Cox PH model

SESSION 3:

• Introduction to the Stratified Cox procedure
• The no-interaction Stratified Cox model
• The Stratified Cox model that allows for interaction

SESSION 4:

• Definition and examples of time-dependent variables
• Definition and features of the extended Cox model
• Stanford Heart Transplant Study Example
• Addicts Dataset Example
• The likelihood function for the extended Cox model.

Dr. Kleinbaum is internationally known for his textbooks in statistical and epidemiologic methods and as an outstanding teacher.  His popular text "Survival Analysis - A Self Learning Text" is the text for this course. He has also taught over 150 short courses over the past 30 years throughout the world.

Prof. Mathew Strickland is Assistant Professor in the Department of Environmental and Occupational Health at Emory University, and will be the primary discussion leader for this session of the course.  He has taught a variety of in-person and distance education courses on Epidemiologic Modeling, Fundamentals of Epidemiology, and Maternal/Child Health Epidemiology. Participants can ask questions and exchange comments with the instructors via a private discussion board throughout the period.

You will be able to ask questions and exchange comments with Dr. David Kleinbaum and Prof. Matthew Strickland via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune.

For India Registration and pricing, please visit us at www.india.statistics.com.

Email: info@c-elt.com

Call: 020 66009116

Websites:

www.india.statistics.com

www.c-elt.com

Introduction to Bayesian Computing and Techniques

Bayesian methods, recently unleashed by computing power, have become very popular.  They answer questions like "given the state of the world, what is the probability that...?"  These are often more useful questions than those starting with "given the null hypothesis, ..."

Get your hands on Bayesian computing with Peter Congdon's online course "An Introduction to Bayesian Computing and Techniques" at Statistics.com. For more details please visit at http://www.statistics.com/BayesianComputing/.

Participants in "An Introduction to Bayesian Computing and Techniques" will learn why Bayesian computing has gained wide popularity, and how to apply Markov Chain Monte Carlo techniques (MCMC) to Bayesian statistical modeling using WinBUGS software. Participants will learn how to use WinBUGS software, use it to estimate parameters of standard distributions, and implement simple regression models.  This course is recommended as a lead-in to "Bayesian Regression Modeling via MCMC," also with Dr. Congdon.

Who Should Take This Course:

Statistical analysts and consultants who need to make decisions (or advise decision-makers) via a process that incorporates domain-specific information -- not simply abstract and arbitrary statistical rules.

Course Program:

Course outline: The course is structured as follows

SESSION 1:

• Basic ideas of MCMC
• Benefits of Bayes methods
• Priors and Prior Informativeness
• Important distributions in Bayesian analysis
• Introduction to three standard schemes: (normal data, normal prior; binomial data, beta prior; poisson data, gamma prior)

SESSION 2:

• Winbugs syntax and programs, data inputs, convergence checks, obtaining summaries

SESSION 3:

• Main elements of posterior summarization
• Tests on parameters or parameter collections (posterior probability tests)
• Model predictions

SESSION 4:

• Simple regression models (normal, binary & binomial, poisson)

Dr. Peter Congdon is a Research Professor in Quantitative Geography and Health Statistics at Queen Mary University of London. He is the author of "Bayesian Statistical Modeling, Applied Bayesian Modeling," and "Bayesian Models for Categorical Data," all published by Wiley, as well as numerous articles in peer-reviewed journals. His research interests include spatial data analysis, Bayesian statistics, latent variable models and epidemiology. Participants can ask questions and exchange comments with Dr. Congdon via a private discussion board throughout the period.

You will be able to ask questions and exchange comments with Dr. Michael Chernick and Dr. Robert LaBudde via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune.

For India Registration and pricing, please visit us at www.india.statistics.com.

Email: info@c-elt.com

Call: 020 66009116

Websites:

www.india.statistics.com

www.c-elt.com

Bootstrap Methods

The bootstrap is now a vital statistical tool and, due to its power and simplicity, its use is ubiquitous.  Learn more in Dr. Chernick's online course "Bootstrap Methods" at Statistics.com. For more details please visit http://www.statistics.com/bootstrap/.

"Bootstrap Methods" covers the basic theory and application of the bootstrap family of procedures, with the emphasis on applications. After taking this course, participants will be able to use the bootstrap procedure to assess bias and variance, test hypotheses, and produce confidence intervals.  The bootstrap is illustrated also for regression and time series procedures. Basic and improved bootstrap procedures are covered.  The software used is mostly R.

Who Should Take This Course:

Statisticians and data analysts who perform statistical inference, or need to assess uncertainty in their data. Those working with data that does not meet the distributional requirements of standard statistical procedures or with unusual statistics or complex estimators will find the course particularly useful.

Course Program:

Course outline: The course is structured as follows

SESSION 1: Introduction

• Wide range of application
• Historical notes
• Bias estimation
• Efron's patch data example
• Estimating other parameters of a distribution

SESSION 2: Parameter Estimation

• Bias estimation (continued)
• Error rate estimation problems
• Confidence intervals and hypothesis test
• Percentile method confidence intervals
• Higher order bootstrap confidence intervals
• A 1-1 relationship between confidence intervals and hypothesis tests
• Problems with bootstrap confidence intervals for variances

SESSION 3: Regression, Time Series, Which Methods?

• Linear Regression, bootstrap residuals or vectors
• Non-linear Regression
• A Quasi-optical experiment
• Nonparametric Regression
• Cox Model
• CART
• Bootstrap Bagging
• Time Series Analysis
• Model-based vs block resampling
• Bootstrap variants
• Bayesian bootstrap
• Smoothed bootstrap
• Parametric bootstrap
• Iterated bootstrap
• Number of repetitions (replications)

SESSION 4: Special Topics, Bootstrap Failures and Remedies

• Spatial data: kriging
• Subset selection
• Examples of Gong and Gunter
• p-value adjustment
• Process capability indices
• Bioequivalence
• Failure Due to Small Sample Size
• Failure Due to Infinite Moments and Remedy (introducing m-out-of-n bootstrap)
• Failure Due to Estimating Extremes and Remedies

Dr. Michael Chernick and Dr. Robert LaBudde are the coauthors of "Bootstrap Methods with Applications in R," the course text.  Chernick studied with Bradley Efron in the bootstrap's early days, is a Fellow of the American Statistical Association, the 1983 winner of the Wolfowitz Prize, and the author of more than 30 journal articles.  LaBudde is president and founder of Least Cost Formulations, Ltd., a mathematical software development company specializing in optimization and process control software for manufacturing companies.

You will be able to ask questions and exchange comments with Dr. Michael Chernick and Dr. Robert LaBudde via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune.

For India Registration and pricing, please visit us at www.india.statistics.com.

Email: info@c-elt.com

Call: 020 66009116

Websites:

www.india.statistics.com

www.c-elt.com

Biostatistics 2

William Weldon, the 19th century British zoologist, was present at the creation of modern biostatistics.  In 1893 he served, with Karl Pearson, on a Royal Committee "For the Purpose of Conducting Statistical Inquiry into the Variability of Organisms."  A year later, he spent considerable time tossing a set of 12 dice more than 26,000 times and recording the results.  Pearson used these data in his paper on the chi-square.

The Institute for Statistics Education at Statistics.com offers courses in a number of topics in biostatistics throughout the year. Statistics.com offers an online courses “Biostatistics 2”. This course covers clinical trial designs including randomized controlled trials, ROC curves, CI and tests for relative risk and odds ratio, and an introduction to survival analysis, and is taught by Dr. Abhaya Indrayan. For more details please visit at http://www.statistics.com/biostat2/.

Who Should Take This Course:

This non-mathematical course is specially designed for medical and health professionals who deal with medical data and want to acquire some statistical skills. These include nursing, pharmacy, laboratory technology and nutrition professionals beside physicians, surgeons and dentists.

Course Program:

Course outline: The course is structured as follows

SESSION 1: Introduction to Designs for Clinical Trials

• Control group
• Randomization and matching
• Blinding and masking
• One-way, two-way and repeated measures designs
• Cross-over design

SESSION 2: ROC Curves

• Review of sensitivity-specificity
• Serial and parallel tests
• ROC curve
• Predictivity-based ROC curve
• Clinical gains from a test

SESSION 3: Relative Risk (RR) and Odds Ratio (OR)

• Definitions and applicability of RR and OR
• CI and test of hypothesis for RR
• CI and test of hypothesis for attributable risk
• CI and test of hypothesis for OR

SESSION 4: Introductory Survival Analysis

• Censored durations and need of special methods
• Life table method
• Kaplan-Meier method

Dr. Abhaya Indrayan, author of "Medical Biostatistics" (3rd ed., CRC Press), has also written a book on medical research methods. He is former Professor and Chair of the Department of Biostatistics and Medical Informatics at Delhi University College of Medical Sciences, Delhi. He has frequently provided biostatistical consultation to the World Health Organization, the World Bank and UNAIDS. Dr Indrayan has delivered the inaugural address at Biostatistics Workshop at Institute of Cytology and Preventive Oncology, NOIDA, been a Guest of Honor at Workshop on Designs and Sample Size at Postgraduate Institute of Medical Sciences and Research, Chandigarh and presented papers in more than 20 international conferences. 

You will be able to ask questions and exchange comments with the Dr. Abhaya Indrayan via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune.

For India Registration and pricing, please visit us at www.india.statistics.com.

Email: info@c-elt.com

Call: 020 66009116

Websites:

www.india.statistics.com

www.c-elt.com