Computing and Data Analysis for Environmental Applications, Fall 2003

Abstract:: Covers computational and data analysis techniques for environmental engineering applications. First third of subject introduces MATLAB and numerical modeling. Second third emphasizes probabilistic concepts used in data analysis. Final third provides experience with statistical methods for analyzing field and laboratory data. Numerical techniques such as Monte Carlo simulation are used to illustrate the effects of variability and sampling. Concepts are illustrated with environmental examples and data sets. This subject is a computer-oriented introduction to probability and data analysis. It is designed to give students the knowledge and practical experience they need to interpret lab and field data. Basic probability concepts are introduced at the outset because they provide a systematic way to describe uncertainty. They form the basis for the analysis of quantitative data in science and engineering. The MATLAB® programming language is used to perform virtual experiments and to analyze real-world data sets, many downloaded from the web. Programming applications include display and assessment of data sets, investigation of hypotheses, and identification of possible casual relationships between variables. This is the first semester that two courses, Computing and Data Analysis for Environmental Applications (1.017) and Uncertainty in Engineering (1.010), are being jointly offered and taught as a single course.

Read the Fine Print

Introduction to Mark-Recapture Census Methods Using the Seed Beetle, Callosobruchus maculatus

Abstract:: Population size, or the abundance of organisms in a study site, is the most fundamental of the primary demographic statistics. We present a laboratory study that introduces college undergraduates to mark-recapture methods that estimate population size. Students apply a simple mark-recapture method to estimate population size in cultures of a seed beetle, Callosobruchus maculatus. After completing this study, students not only will have rudimentary knowledge of statistical methods (e.g., standard deviation and 95% confidence limits), but will also know how to obtain reliable estimates of population size.

Abstract:: Studies how randomization can be used to make algorithms simpler and more efficient via random sampling, random selection of witnesses, symmetry breaking, and Markov chains. Models of randomized computation. Data structures: hash tables, and skip lists. Graph algorithms: minimum spanning trees, shortest paths, and minimum cuts. Geometric algorithms: convex hulls, linear programming in fixed or arbitrary dimension. Approximate counting; parallel algorithms; online algorithms; derandomization techniques; and tools for probabilistic analysis of algorithms.

Abstract:: Statistical Reasoning in Public Health provides an introduction to selected important topics in biostatistical concepts and reasoning through lectures, exercises, and bulletin board discussions. It represents an introduction to the field and provides a survey of data and data types. Specific topics include tools for describing central tendency and variability in data; methods for performing inference on population means and proportions via sample data; statistical hypothesis testing and its application to group comparisons; issues of power and sample size in study designs; and random sample and other study types. While there are some formulae and computational elements to the course, the emphasis is on interpretation and concepts.

Subject:: Mathematics and Statistics, Science and Technology, Social Sciences
Grade Level:: Post-secondary
Collection:: JHSPH OpenCourseWare