Training course "Introduction to descriptive and parametric statistic with R"
(Course no. 98/2016 in the training programme of Forschungszentrum Jülich)
Target audience: | Staff members who want to use R for statistical analysis |
Contents: | |
Language: | This course is taught in English. |
Duration: | 1 day |
Date: | Thursday, 14 April 2016, 9:00-12:00 and 13:30-15:00 |
Venue: | Jülich Supercomputing Centre, Besprechungsraum 2, building 16.3, room 315 |
Number of participants: | minimum 5, maximum 14 |
Instructors: | Dr. Antoine Tordeux, JSC |
Contact: | Dr. Antoine Tordeux |
Some basic statistical tools are presented and investigated using R freeware and data from pedestrian trajectories.
The objective is to propose useful methods allowing to analyze data or to calibrate a model, as well as to learn how to use R.
The course is organized in three sessions:
- The first session is about basic descriptive tools for univariate and bivariate data (such as histogram, kernel density, measures of location/variability, boxplot, covariance, correlation, simple linear regression, qqplot, Bootstrap methods).
- The second session concerns descriptive analysis of multivariate data: Principal component analysis and Multiple linear regression / Estimation by least-square.
- The last session is about parametric statistic: Estimator (definition and property); Punctual estimation of parameters (by maximal likelihood and Bayesian approach); Information criteria (AIC, BIC, likelihood ratio...); Precision of the estimations (confidence and credible intervals); Tests of hypothesis.
Participants are requested to bring their own laptops.