Tutorials - Statistical Methods

Subject Components Tutorial Dataset
Descriptive Statistics
Some descriptive statistics on an dataset.
Data importation wizard
Dataset
Define Status
Univariate continuous stat
Univariate discrete stat
Group characterization
breast
Correlation Coefficient
Linear Correlation Coefficient. Significance testing. Partial Correlation.
Dataset
Define Status
Linear Correlation
Residual Scores
cars
Measures of association for nominal variables
Contingency/crosstabulation table, Chi-square test of independence, PRE measures.
Dataset
Define Status
Linear Correlation
Goodman Kruskal Lambda
Goodman Kruskal Tau
Theil U
cars consumption
Association measure for ordinal variables
Contingency table, pairwise comparison, concordant and discordant pairs : Goodman and Kruskal's Gamma, Kendall's Tau-b and Tau-c de Kendall, Sommers' d.
Dataset
Define Status
Linear Correlation
Goodman Kruskal Gamma
Kendal tau-b
Kendal tau-c
Sommers d
blood pressure
Nonparametric Statistics
Using nonparametric statistics for studying household behavior, including: Mann-Whitney, Kruskal-Wallis, Spearman's rho, Kendall's tau, sign test, Wilcoxon matched-samples test
Nonparametric Statistics
Test
nonparametric statistics data
ANOVA -- Test of equality of variances
ANOVA : compare gear diameter from various batches
Tests for equality of variances.
ANOVA
Bartlett's / Levene's / Brown & Forsythe's tests for equality of variances
GEAR dataset (from NIST)
Two-sample T-Test
Two-sample T-TEST for Equal mean: to compare the consumption of vehicles according to their origin.
T-Test Equal & Unequal Variances
AUTO83B dataset (from NIST)
Normality Test
Goodness-of-fit tests for checking the normality of the distribution. We use artificial dataset in this tutorial, we have generated the dataset from 3 distributions : uniform, normal and log-normal.
Normality Test
Dataset
MANOVA
One Way Manova -- Multivariate Analysis of Variance : unlike classical ANOVA, there is more than one dependent variables.
Comparing the results with ANOVA on each dependent variable.
One Way MANOVA
Manova Dataset
Friedman's Test
Friedman's Two-Way Analysis of Variance by Ranks
Comparing the performances of lecturers according to their use of slides.
Friedman's ANOVA by Ranks
Lecturer performances
Partial Correlation
Computing the correlation between two variables by controlling the effect of other variables.
Partial Correlation
WAIS test
Semi-partial Correlation
Computing the explanation of a dependent variable by an independent variable, by controlling the effect of other variables on this last one.
Semi-partial Correlation
cars


Ricco Rakotomalala.