Multivariate Statistical Analysis
It can be difficult to know what statistical analysis to use when your research question involves a suite of correlated observed variables. This often happens when you have many measured variables as your outcome, or when your set of predictor variables are not fully independent. Some studies consider how multiple predictors explain multiple response variables. The term ?multivariate statistics? includes a myriad of techniques designed to address the above situations. This workshop will provide a broad overview of some of the most commonly used multivariate analyses, including both descriptive and inferential approaches. Topics will include:
- Dimension reduction techniques, including:
- Principle Components Analysis
- Redundancy Analysis
- Non-metric Multidimensional Scaling (NMDS)
- Clustering techniques, including:
- Hierarchical and k-means
- Hypothesis testing, including:
- Mantel Tests
- Permutation techniques
Fee: None to members of the Cornell community, but registration is required. Since space is limited, early registration is encouraged.
For times and locations of upcoming workshops, please see the Workshop Schedule.