1.40814 NON-PARAMETRIC STATISTICS

Parametric and non-parametric tests. Advantages and disadvantages of non-parametric tests over parametric tests. Probability integral transformation, empirical distribution function. Different non-parametric tests. Chi-square goodness of fit tests, sign tests, Wilcoxon tests, Wilcoxon signed-rank test, Wilcoxon rank-sum test, test of randomness, run test, median test, Mann-Whitney U-test, Kolmogorov-Smirnov tests (both one and two sample problems), Spearman's rank correlation and test, Kruskal-Wallis test, Friedman test etc.

Contact hours:

Four lectures/tutorials and one computer lab per week.

Assessment:

Two tests (25%), approximately four assignments (15%), final exam (60%).

Prerequisite: 1.30812

Text:

Siegel, S. & Castellan, N. J., 1956, Non-parametric Statistics, McGraw-Hill, N.Y.

Walsh, J. E., 1962, Handbook of Non-Parametric Statistics, D. Van Nostrand

Triola, M. F., 1989, Elementary Statistics, Benjamin/Cummings, N.Y.