Statistics for Engineers and Life Scientists

NCSU courses ST311, ST361 and ST371

Over the past couple of decades there has been unprecedented efforts to obtain data in almost all fields of sciences and engineering. This requires that we provide training of basic statistical and applied mathematical methods that are necessary for handling such vast amounts of data that would enable any practitioners to extract `signals’ by separating the `noises’ from vast amount of data. Knowledge of modern data science methods equipped with computational techniques are becoming necessary tools for any scientists ranging from astronomy to zoology. Statistical methods provide systematic approaches for describing and interpreting information so that we can make the most informed decisions in practice. The proposed course aims to equip students with basic ideas and necessary computational tools in statistics, which range from systematic data collection to making probabilistic inference and decisions. The course will begin with a collection of introductory concepts and commonly used methods in statistics, illustrate the use of modern computational approaches to statistics by introducing the use a commonly used software for modern data science. Students will be working in teams to exploring ideas on data examples and case studies using the statistical software R. The course may not have time to cover all theoretical underpinnings and advanced mathematical tools but the students would have access to a multitude of online materials to gain further insights to understand the basic technical methods of mathematical statistics. The course will cover the following topics in class:

1. Basic Probability Theory (brief review)

2. Data Collection and Visualization (using R)

3. Sampling Distributions (using Monte Carlo methods)

4. Point and Interval Estimation Methods

5. Statistical Hypothesis testing

6. Regression Methods (multiple linear regression, logistic regression and Poisson regression)

7. Statistical Model diagnostics, Predictive inference and Goodness-of-fit methods

8. Optional Topics: Maximum likelihood principle and Elementary Nonparametric methods

Prerequisite: Students are expected to be fluent with elementary level quantitative probabilistic reasoning and analysis (for example, conditional probability, independence of events, random variables, univariate and multivariate probability distributions, correlation and covariance). Some of the course assignments will involve coding and statistical analysis on datasets provided. It is recommended that students familiarize themselves with the use R for this purpose using online resources provided in class.