Nonparametric Statistical Methods

In statistics, there is no one, perfect test, i.e. a test that would be optimal in every possible situation. Each test is based on certain preconditions. Traditionally, parametric tests are used exclusively, believing that the schedule of the group is, if not completely, than at least approximately normal. However, this assumption in practice is not always realistic, so nonparametric tests are often used, and since they do not depend on the schedule of the group, an alternative name is used: tests independent of the of the group.

It consists in mastering of the candidate/student with statistical methodology and its application, so that in various analyzes and research procedures they could obtain results on the basis of which, in future, valid conclusions and appropriate decisions could be derived.

Application of nonparametric statistical methods; Test of the sign and the confidence interval; Normal approximation; Mann-Whitney U-test; Wald-Wolfowitz test: Wilcoxon rank sum test; Kruskal-Wallis analysis of variance ranks (H-test); The median test; Freedman analysis of variance; Spearman rank correlation; Practical application of tests in SPSS software.

Lectures and exercises, with the seminary classes.

Colloquium I (0-20 points); Colloquium II (0-20 points); Final exam (0-50 points); Attendance (2point); Class activities (seminar papers, discussions, case studies) (0-8 points)

Zlatko Kovačić (1998). Multivariate analysis. Belgrade: Faculty of Economics; Manly, B. F. J. (1991). Multivariate Statistical Methods - A Primer. London: Chapman and Hall; Mardia, K. V.; Kent, J. T .; Bibby, J. M. (1979); Multivariate Analysis. London: Academic Press; SPSS (2003). SPSS Advanced Models 12.0. Chicago: SPSS Inc .; Newbold, P .; Carlson, W. L .; Thorne, B. (2010). Statistics for Business and Economy. Zagreb: Mate; Lovrić, M.; Komić, J .; Stević, S. (2006). Statistical analysis - methods and applications. Banja Luka: Faculty of Economics.