4:00pm to 6:00pm |
|
LISA Statistics Short Course: Multivariate Analysis in R
(Academic)
LISA SHORT COURSES IN STATISTICS
LISA (Virginia Tech's Laboratory for Interdisciplinary Statistical Analysis) is providing a series of evening short courses to help graduate students use statistics in their research. The focus of these two-hour courses is on teaching practical statistical techniques for analyzing or collecting data. See www.lisa.stat.vt.edu/?q=short_courses for instructions on how to REGISTER and to learn more.
Spring 2015 Schedule:
Monday & Tuesday, February 16 & 17: Basics of R;*
Monday & Tuesday, February 23 & 24: Graphics in R;*
Tuesday, March 3: Multivariate Analysis in R;
Tuesday, March 17: Designing Experiments;
Monday & Tuesday, March 23 & 24: Using ggplot2 to produce enhanced graphics in R;*
Tuesday, April 7: T-tests & ANOVA;
Tuesday, April 14: Solutions for Broken Linear Models;
*Two sessions to accommodate more attendees.
Tuesday, March 3;
Instructor: Liang (Sally) Shan;
Title: Multivariate Analysis in R;
Course Information:
Multivariate analysis is commonly used when we have more than one outcome variables for each observation. For instance, a survey of American adults' physical and mental health might measure each person's height, weight, and IQ. In this scenario, the three outcome variables are measured simultaneously, and you may expect some extent of correlation among the outcome variables (e.g., A taller person may also has a heavier weight). The primary goal of this short course is to help researchers with multivariate data better visualize and understand their data using multivariate analysis tools.
In this course, we will focus on dimension reduction techniques that help reduce the number of variables. We will be mainly talking about two techniques, Principle Component Analysis (PCA) and factor analysis. Note that dimension reduction is different from clustering, where the latter involves methods to place observations into groups. R software will be used in this course.
This course covers:
1. Differences between multivariate analysis and univariate analysis
2. Differences between dimension reduction and clustering
3. Principle Component Analysis (PCA)
4. Factor analysis
5. Relationship between PCA and Factor Analysis
Data Set:
The Iris flower data set will be used for illustration purpose. It includes 50 samples from each of three species of Iris (setosa, virginica and versicolor). Four outcome variables were measured from each sample: the length and the width of the sepals and petals.
The graph below shows the PCA analysis result of the Iris flower data. We will explore and explain more about it in our short course.
www.lisa.stat.vt.edu/sites/default/files/MultivariateStatisticsinR.png
Follow us on Facebook (www.facebook.com/Statistical.collaboration) or Twitter (www.twitter.com/LISA_VT) to be the first to know about LISA events! More information...
|