StatPlus
Library
Online Courses in Statistics by Statistics.com
Adaptive/Group Sequential Designs for Clinical Trials -This course will teach you how to design, monitor and analyze clinical trialsusing statistically sound principles that incorporate interim looks at thedata, possible early stopping, and interim re-estimation of power andrequired sample size. It covers group sequential designs and adaptive methodsof sample-size re-estimation.
Advanced Design of Experiments - The aim of the course isto present advanced and important concepts that have received very littleattention, such as designs for irregular experimental regions and Analysis ofMeans (ANOM).
Advanced Resampling Methods - The course extends therange of application of decision-free procedures including the bootstrap,decision trees and permutation (randomization) tests. All sessions includecritical appraisal methodology and exercises. Participants will learn toanalyze experimental designs, create their own statistics, analyzecategorical data as well as combinations of categorical and continuous data,and develop and validate models.
Basic Concepts in Probability and Statistics - Thiscourse provides an easy introduction to statistics and statisticalterminology through a series of practical applications. Once you've completedthis course you'll be able to summarize data and interpret reports andnewspaper accounts that use statistics and probability. You'll use simulationand resampling to fully grasp the difficult concept of "statisticalsignificance."
Introduction to Bayesian Statistics - This course willintroduce you to the basic ideas of Bayesian Statistics. You will learn howto perform Bayesian analysis for a binomial proportion, a normal mean, thedifference between normal means, the difference between proportions, and fora simple linear regression model.
Introduction to Biostatistics - This course covers theprincipal statistical concepts used in biostatistics. Basic concepts commonto all statistical analysis are reviewed, and those concepts with specificimportance in biostatistical are covered in detail.
Categorical Data Analysis - This course will cover theanalysis of contingency table data (tabular data in which the cell entriesrepresent counts of subjects or items falling into certain categories). Topics include tests for independence (comparing proportions as well as chi-square), exact methods, and treatment of ordered data. Both 2-way and 3-way tables are covered.
Clinical Trial Design - This course covers the essential concepts required to design rigorous randomized trials so as to ensure valid treatment comparisons.
Cluster Anaylsis - This course will teach you how to use various cluster analysis methods to identify possible clusters in multivariate data. Methods discussed include hierarchical clustering, k-means clustering, two-step clustering, and normal mixture models for continuous variables.
Introduction to Data Mining - This course covers the two core paradigms that account for most business applications of data mining: classification and prediction. In both cases, data mining takes data where avariable of interest is known and develops a model that relates this variableto a series of predictor variables. In classification, the variable ofinterest is categorical ("purchased something" vs. "has not purchasedanything"). In prediction, the variable of interest is continuous ("dollarsspent"). Four techniques will be used: k-nearest neighbors, classificationand regression trees (CART), logistic regression and multiple linearregression. The course will also cover the use of partitioning to divide thedata into training data (data used to build a model), validation data (dataused to assess the performance of different models, or, in some cases, tofine tune the model) and test data (data used to predict the performance ofthe final model).
Data Mining: Unsupervised Techniques - This course coverskey unsupervised learning techniques association rules, principalcomponents analysis, and clustering. The course will include an integrationof supervised and unsupervised learning techniques.
Decision Trees and Rule-Based Segmentation - Ruleinduction is an important component of data mining, and this course coverstwo main styles of generating rules.
Design of Experiments - This course will stress theapplication of DOE rather than statistical theory. With a 12-step checklist,it covers full and fractional factorial designs, Plackett-Burman,Box-Behnken, Box-Wilson and Teguchi designs.
Directional (Circular) Data - Directional data (alsocalled circular data) are data that are measured on a scale that repeatsitself hours in the day or angular directions are prime examples. Thiscourse will cover the exploratory and inferential tools needed to analyzesuch data and course participants will gain hands on software experience.
Statistics for Engineers - This course covers topics instatistics that are of special concern to engineers. Topics covered includeprediction intervals, tolerance intervals, calibration intervals, measurementerror, accelerated life testing, measurement system appraisal, reliabilityand lifetime testing.
The Statistics of Environmental Impact Assessment - Thiscourse will introduce you to the statistical methods used in environmentalanalysis. Many of these methods would be covered in a standard course onstatistics, but some of the topics that are covered here would not beincluded in such a course.
Fundamentals of Epidemiology - This is an introductoryepidemiology course that emphasizes the underlying concepts and methods ofepidemiology. Topics covered in the course include: study designs (clinicaltrials, cohort studies, case-control studies, and cross-sectional), measuresof disease frequency and effect.
Bias in Epidemiologic Research - This is a second levelepidemiology course that emphasizes the underlying concepts and methods foraddressing validity and bias issues in epidemiologic research. Topics coveredin the course include: overview of validity and bias, selection bias,information bias, and confounding bias.
Analysis of Epidemiologic Data - This is a second levelepidemiology course that emphasizes methods for analyzing epidemiologic data.Topics covered in the course include: simple analysis of 2x2 tables, controlof extraneous variables (including an introduction to logistic regression),stratified analysis, and matching.
Exploration and Analysis of DNA Microarray Data - Thiscourse will acquaint you with the process of microarray data mining frombeginning to end. You will learn how to how to preprocess the data, estimategene expression patterns, cluster genes to detect interesting gene expressionpatterns, and classify experiments (subjects) based on gene expressionpatterns. Illustrations of the statistical issues involved at the variousstages of the analysis will use real data sets from DNA microarrayexperiments.
Mixed Effects Models with Applications - This course willexplain the basic theory of linear and non-linear mixed effects models. Itwill outline the algorithms used for estimation, primarily for modelsinvolving normally distributed errors, and will provide examples of dataanalysis. The course aims at providing a basic understanding and knowledge ofthe mixed effect models that will allow you to use them in practice.
Financial Risk Management - Modeling Derivatives - Thiscourse introduces basic stochastic models for financial derivatives such asoptions and futures -- important instruments in risk management. The coursecombines theoretical and practical aspects of option pricing and trading,using real world examples for illustration, and focuses on discrete timemodels for option pricing and trading.
Generalized Linear Models (GLM) - This course willexplain the theory of generalized linear models (GLM), outline the algorithmsused for GLM estimation, and explain how to determine which algorithm to usefor a given data analysis. GLM allows the modeling of responses, or dependentvariables, that take the form of counts, proportions, dichotomies (1/0),positive continuous values, as well as values that follow the normal Gaussiandistribution. Logistic, Poisson, and negative binomial regression models arethree of the most noteworthy GLM family members.
Latent Variable Growth Curve Modeling - This course willintroduce you to the topic of latent variable growth curve modeling, whichtakes traditional modeling of growth curves for repeated measures data, andextends it to cover the use of latent variables via structural equationmodeling (SEM) methods.
Logistic Regression - Logistic regression extendsordinary least squares (OLS) methods to model data with binary (yes/no,success/failure) outcomes. Rather than directly estimating the value of theoutcome, logistic regression allows you to estimate the probability of asuccess or failure.
Meta Analysis - This course will explain meta analysis -the methods that are used to assess multiple statistical studies on the samesubject and draw conclusions.
Modeling Count Data - This course deals with regressionmodels for count data; i.e. models with a response or dependent variable datain the form of a count or rate. The course will cover Poisson regression, thefoundation for modeling counts, as well as extensions and modifications tothe basic model.
Modeling Longitudinal and Panel Data - This course coversthe extension of Generalized Linear Models (GLM) to model varieties oflongitudinal and clustered data, called panel data.
Queueing Theory - This course provides a firm foundationin queueing analysis, optimization of queues, and design of queueing systems.Although seemingly abstract, this subject is actually extremely practical asqueueing experts are demanded throughout the world for such tasks as serverdesign through traffic analysis, aircraft and vehicle traffic flow, inboundcall management, and a variety of other applications.
Introduction to R - This 3-week course will provide aneasy introduction to R and its use in statistics and in organizing data. Onceyou've completed this course you'll be able to enter, save, retrieve,summarize and display data, run simulations, and test hypotheses using R.
Modeling in R and S-PLUS - This 3-week course will showyou how to use R and S-PLUS to create models for use in classification andprediction. You will be introduced to advanced graphing methods as needed.Modeling techniques include OLS, LAD, and EIV regression, quantileregression, and decision trees.
Practical Rasch Measurement - Rasch analysis constructslinear measures from scored observations, such as responses tomultiple-choice questions, Likert scales and quality-of-life assessments.This course covers the practical aspects of data setup, analysis, outputinterpretation, fit analysis, differential item functioning, dimensionalityand reporting.
Real Estate Pricing and Financial Stability - This coursecovers the statistical methodologies used in constructing both commercial andresidential real estate price indexes, which are important tools thatfinancial institutions can use to monitor their exposure to risk fromvolatility in real estate markets. It also addresses relationships betweenreal estate prices and banking profitability, and the roles that bank credit,GDP, stock equity prices and interest rates play in determining real estateprices.
Introduction to Resampling Methods - This internet coursecourse introduces the basic concepts and methods of statistics via resamplingmethods. Participants will use Resampling Stats, S-PLUS or R (depending onpreference; Resampling Stats is recommended for those unfamiliar with S-PLUSor R) to do interval estimation, one- , two- and k-sample comparisons,correlation, and a number of other most-powerful statistical procedures. Thegoal of the course is to give participants the confidence and tools necessaryfor the practice of statistics in their own research and in interpreting theresearch of others. Taught by Dr. Philip Good, author of "Resampling Methods"(Birkhauser) and "Permutation Tests" (Springer).
Introduction to Regression - Regression, perhaps the mostwidely used statistical technique, estimates linear relationships betweenindependent (explanatory) variables and a dependent (outcome) variable.Regression models can be used to help understand and explain relationshipsamong variables; they can also be used to predict actual outcomes. In thiscourse you will learn how regression models are derived, use software toimplement them, learn what assumptions underlie the standard regressionmodel, learn how to test whether your data meet those standard assumptions,and learn what can be done when those assumptions are not met.
Sample Size and Power Determination - This course showsyou how to make power and sample size determination for experiments, surveysand long-term trials.
Spatial Statistical Analysis in Geographic Information Systems(GIS) - Spatial statistical analysis uses methods adapted fromconventional statistics to address problems in which spatial location is themost important explanatory variable. This course will explain and giveexamples of the analysis that can be conducted in a geographic informationsystem such as ArcGIS or Mapinfo.
Statistical Process Control (With Applications in the HealthServices) - This course will explain the theory and practice ofusing control charts to monitor and control processes with an emphasis onapplication in the health service area.
Introduction to Statistics I: Inference for a SingleVariable - The aim of this course is to provide an easy introductionto statistics and statistical terminology through a series of practicalapplications. Once you've completed this course you'll be able to testhypotheses regarding proportions and means. You'll use simulation andresampling to fully grasp the difficult concept of "statisticalsignificance."
Introduction to Statistics II: Working with BivariateData - The aim of this course is to provide an easy introduction toinference for two variables through a series of practical applications. Onceyou've completed this course you'll be able to test hypotheses regarding asimple regression or a comparison of proportions or two means.
Structural Equation Modeling - This course covers thefundamental concepts and theory of Structural Equation Modeling -- describingthe relationships between variables. Case studies are used and AMOS softwareis introduced.
Advanced Structural Equation Modeling - This coursecovers many popular advanced SEM models with practical exercises. Modelscovered include Multiple Indicator an Multiple Causes models (MIMIC),Multiple Group models, Multilevel (HLM) models, Mixture models, StructuredMeans models, Multitrait-Multimethod models, Second Order Factor models,Interaction models, and Dynamic Factor models.
Survey Design and Sampling Procedures - This coursecovers the crafting of survey questions, the design of surveys, and differentsampling procedures that are used in practice. Longstanding basic principlesof survey design are covered, and the impact of the trend toward increasedrespondent resistance is discussed.
Survey Analysis - This course covers the analysis of datagathered in surveys.
Survival Analysis - The course describes the variousmethods used for modeling and evaluating survival data, or time-to eventdata.
Introduction to S-PLUS - This course should get youstarted using the S-PLUS statistical package and understanding how to writeS-PLUS script programs. Topics include basic statistical analysis, trellisgraphing, hypothesis testing, Monte Carlo simulation, cross-validation,bootstrap, jackknife, and meta-analysis.
Text Mining - This course will introduce the essentialtechniques of text mining, understood here as the extension of dataminings standard predictive methods to unstructured text.
Time Series Forcasting - This course will teach you howto choose an appropriate time series model, fit the model, to conductdiagnostics, and use the model for forecasting. The course will focus onAutoregressive (AR), Moving Average (MA), combined ARMA, and Box Jenkins typemodels.
Toxicological Risk Assessment - This course will coverthe statistical procedures used in analyzing the risk from toxic substances,primarily the results of experiments. It covers significance tests for trendsand their application to dose response relationships, modeling techniques fordose response relationships, benchmark dose estimation, and the incorporationof historical control information in dose response modeling.
