# Practical Statistical Methods

## A SAS Programming Approach

304 pages | 27 B/W Illus.

Hardback: 9781439812822
pub: 2011-04-25
US Dollars\$97.95
x

Practical Statistical Methods: A SAS Programming Approach presents a broad spectrum of statistical methods useful for researchers without an extensive statistical background. In addition to nonparametric methods, it covers methods for discrete and continuous data. Omitting mathematical details and complicated formulae, the text provides SAS programs to carry out the necessary analyses and draw appropriate inferences for common statistical problems.

After introducing fundamental statistical concepts, the author describes methods used for quantitative data and continuous data following normal and nonnormal distributions. She then focuses on regression methodology, highlighting simple linear regression, logistic regression, and the proportional hazards model. The final chapter briefly discusses such miscellaneous topics as propensity scores, misclassification errors, interim analysis, conditional power, bootstrap, and jackknife.

With SAS code and output integrated throughout, this book shows how to interpret data using SAS and illustrates the many statistical methods available for tackling problems in a range of fields, including the pharmaceutical industry and the social sciences.

Introduction

Types of Data

Descriptive Statistics/Data Summaries

Graphical and Tabular Representation

Population and Sample

Estimation and Testing Hypothesis

Normal Distribution

Nonparametric Methods

Some Useful Concepts

Qualitative Data

One Sample

Two Independent Samples

Paired Two Samples

k Independent Samples

Cochran’s Test

Ordinal Data

Continuous Normal Data

One Sample

Two Samples

k Independent Samples

Multivariate Methods

Multifactor ANOVA

Variance Components

Split Plot Designs

Latin Square Design

Two Treatment Crossover Design

Nonparametric Methods

One Sample

Two Samples

k Samples

Transformations

Friedman Test

Association Measures

Censored Data

Regression

Simple Regression

Polynomial Regression

Multiple Regressions

Diagnostics

Weighted Regression

Logistic Regression

Poisson Regression

Robust Regression

Nonlinear Regression

Piecewise Regression

Accelerated Failure Time (AFT) Model

Cox Regression

Parallelism of Regression Equations

Variance-Stabilizing Transformations

Ridge Regression

Local Regression (LOESS)

Mixture Designs and Their Analysis

Analysis of Longitudinal Data: Mixed Models

Miscellaneous Topics

Missing Data

Diagnostic Errors and Human Behavior

Density Estimation

Robust Estimators

Jackknife Estimators

Bootstrap Method

Propensity Scores

Interim Analysis and Stopping Rules

Microarrays and Multiple Testing

Stability of Products

Group Testing

Correspondence Analysis

Classification Regression Trees (CARTs)

Multidimensional Scaling

Path Analysis

Choice-Based Conjoint Analysis

Meta-Analysis

References and Selected Bibliography

Index