This course unit aims to extend the knowledge acquired in Statistics II, as well as provide students with theoretical and practical knowledge concerning nonparametric statistical inference methodologies. The syllabus of the course comprises instruments nonparametric statistical inference as orders, statistics, order statistics, the estimators and sampling distributions, estimation nonparametric point and interval hypothesis tests and nonparametric. Are treated equally essential aspects of asymptotic distributions. The students should get to know the estimators and their properties, to construct confidence intervals and nonparametric testing of nonparametric hypotheses. For each non-parametric procedure, students should know clearly its applicability conditions.
Tiago André Gonçalves Félix de Oliveira
Weekly - Available soon
Total - Available soon
Portuguese. If there are Erasmus students, classes will be taught in English
Conover, W. (1999). Practical Nonparametric Statistics. John Wiley.; Lehmann, E. (1998). Nonparametrics. Statistical Methods Based on Ranks. Prentice Hall.; Siegel, S. & Castellan, N. (1988). Nonparametric Statistics for the Behavioral Sciences. McGraw-Hill.; Galvão de Mello, F. (1997). Probabilidades e Estatística II. Escolar Editora.; Murteira, B.; Silva Ribeiro, C.; Andrade e Silva,
Theoretical and practical classes and laboratory.
1st Period – Assessment is continuous through three tests. Each test has a weight of 1/3 in the final grate.
2nd Period – Exam (100%).
Statistical Inference. Nominal scale, ordinal scale, interval scale and ratio scale. Populations, samples, parameters, statistics, order statistic. Parametric methods, robust methods and nonparametric methods. Confidence intervals. Hypothesis testing. P-value.
2. ANALYSIS OF A POPULATION
Confidence interval and the quantile test of a population
3. COMPARISON OF TWO POPULATIONS
Independent samples and paired samples
Confidence interval and Hodges-Lehmann estimate of the difference in location of two distributions
4. COMPARISON OF MORE THAN TWO POPULATIONS
Test for variances
Independent or paired samples
Kruskal-Wallis test. Multiple comparisons
Chi-squared test for differences in probabilities
5. RANDOMNESS AND INDEPENDENCE. MEASURES AND ASSOCIATION TESTS
Chi-squared test for independence. Carmér’s contingency coefficient. Pearson’s contingency coefficient
Spearman correlation coefficient and e de Kendall correlation coefficient
6. GOODNESS-OF–FIT TESTS
Lilliefors test. Shapiro-Wilk test.