# Statistics for life and health sciences EBook

Welcome to the UCLA Statistics for the Biomedical and Health Sciences (Stats 13) electronic book (EBook).

## Preface

This is an Internet-based probability and statistics for biomedical and health sciences EBook. The materials, tools and demonstrations presented in this EBook would are used for the UCLA Statistics 13 course. The EBook is developed, updated and manages by the UCLA Statistics faculty teaching this course over the years. Many other instructors, researchers, students and educators have contributed to this EBook.

There are four novel features of this Statistics EBook. It is community-built and allows easy modifications and customizations, completely open-access (in terms of use and contributions), blends information technology, scientific techniques, heterogeneous data and modern pedagogical concepts, and is multilingual.

### Format

Each section in this EBook includes

• Motivation
• Concepts, definitions, formulations
• Examples
• Small (mock-up) and real (research-derived) data
• Webapp demonstration with real data (HTML5)
• R programming
• Problems

### Pedagogical Use

...

The Probability and Statistics EBook is a freely and openly accessible electronic book for the entire community under CC-BY license ...

## Chapter II: Data and variability

• Data
• R data management (Import and Export)
• Histograms, densities and summary statistics

## Chapter III: Randomization-based statistical inference

• Samples, Populations, Repeated Samples, Resampling
• Bootstrapping
• Testing 1, 2 or more samples
• Confidence intervals

## Chapter VI: Parametric Model-based Inference

• Hypothesis testing foundations
• Type I and II errors, Power, sensitivity, specificity

### One sample inference

• T-Test
• Normal Z-test
• Confidence intervals

### Two sample inference

• Independent samples
• Paired samples

• CLT
• LLN

## Chapter VIII: Linear Modeling

• Parametric (simple and multivatiate) regression
• Parametric ANOVA
• Parametric assumptions and model validation
• Non-parametric linear modeling
• Randomization and Resampling based multivariate inference