Syllabus
Overview of basic concepts
Descriptive and mathematical (inductive) statistics. The most common statistical parameters. Normal distribution and the central limit theorem. T-, Chi-squared- and F-distributions. Sampling and statistical independence, correlation.
Basic tests, confidence intervals
The z-test and various types of t-test, the F-test for equality of variances. One- and two-sided tests, ANOVA for one and more factors. Decomposition of variability using different types of sums of squares. Confidence intervals and their relation to hypothesis tests.
Regression models
Their purpose and ways to use them. Linear regression and logistic regression. Multivariate models. Survival analysis, Cox models.
Tests for categorical (qualitative) variables
Chi-squared test of independence (Pearson‘s chi-squared test). Dummy coding for linear and logistic regression.
Pairing
Paired tests, randomization, (randomized) blocking.
Nonparametric methods
The usage of nonparametric methods. Permutation tests, Fisher‘s exact test, the rank-sum test (Mann-Whitney U-test), Wilcoxon‘s test, the Kruskal-Wallis test and the Friedland test.
Classification tasks
Clustering, (linear) discriminant analysis.
Planning of experiments
Design of Experiments (DoE), factorial designs, PCA, statistical power.
Processing of results
Verifying the assumptions, exclusion of outliers, missing values, meta-analysis.
Annotation
The course Basic statistics for doctoral students addresses traditional and modern statistical methods popular in pharmaceutical research. Experience with basic statistical procedures and resulting interpretation of results is expected.