1.30813 ESTIMATION & TEST OF HYPOTHESIS

Point and interval estimation of mean, variance and proportion. Confidence intervals for difference between two population means, difference between two proportions and ratio of two variances. Tests of hypothesis concerning a population mean, proportion and variance. Tests comparing two means, proportions, and variances. Chi-square tests, multinomial experiments, contingency tables, goodness of fit tests. Predictions about the populations based on samples. Non-parametric tests: sign test, run test, Wilcoxon tests, test of randomness, Kolmogorov-Smirnov tests (both one and two sample problems), Spearman's rank correlation test, Kruskal-Wallis test, Friedman test etc. Computer implementations using appropriate statistical software and calculators. This course is designed especially for students from other disciplines.

Contact hours:

4 lectures/tutorials, 1 computer lab per week.

Assessment:

Two tests (25%), approx. 4 assignments (15%) and final exam (60%).

Prerequisite:

1.20811 or 1.20812

Text:

Freund, J. E. & Walpole, R. E. 1984, Mathematical Statistics, Prentice-Hall, N.Y.

Triola, B. C., 1989, Elementary Statistics, 4th ed., Benjamin Cummings, New York