<p>(Formerly numbered 212.) Lecture, three hours; discussion, one hour. Requisite: course 100A. Elements of statistical learning. Topics include linear regression, classification (logistic regression, discriminant analysis), resampling methods (cross-validation, bootstrap), model selection (subset, stepwise) and regularization (ridge, lasso). Letter grading.</p>

Biostatistics

Jonathan and Karin Fielding School of Public Health

Container 1

BIOSTAT 212A - Statistical Learning A

Statistical Learning A

MPH5: Prepare written and oral presentations providing public health insights based on statistical analyses

MS5: Communicate results of biostatistical research both orally and in writing

MS2: Examine foundations of linear and generalized linear statistical models.

Assess the scientific or public health objectives of a data analysis and recommend a modeling strategy to meet the objectives.

MPH2: Employ computational methods for analysis of public data sets.

In homework assignments, students will practice R solutions to various statistical learning tasks on biomedical and public health data sets.

MPH1: Demonstrate mastery of fundamental concepts of statistical analysis for datasets from health studies.

Demonstrate understanding of the theory of linear regression, classification methods (logistic regression, discriminant analysis), resampling methods (cross-validation, bootstrap), linear model selection and regularization.

MS3: Employ computational methods of applied regression to analysis of biomedical data sets

Construct and diagnose linear regression, logistic regression, discriminant analysis, model selection and regularization in R.

Prepare a report synthesizing the results of analysis

General Catalog

Statistical Learning A

BIOSTAT 212A

4 units