Multivariate Analysis
My first attempt at multivariate analysis was during a research project aimed to determine what factors predict whether a physical therapy patient will I initially ran a univariate analysis then only included significant univariate predictors in the final multivariate analysis.
Determine the type of multivariate analysis
layout ### Multivariate Analysis of Variance (MANOVA): “Used when there are multiple dependent variables and the predictors are categorical. Tests whether the means of multiple dependent variables differ across groups.”
Principal Component Analysis (PCA):
“A dimensionality reduction technique. Focuses on finding patterns in data and reducing its complexity, not prediction.”
Structural Equation Modeling (SEM):
“Models complex relationships between multiple dependent and independent variables, including latent variables. Combines aspects of factor analysis and regression.”
Cluster Analysis:
“Groups observations based on similarity without specifying dependent variables. Used in exploratory data analysis.”
Mixed-Effects Models:
“Extends GLMs to include both fixed and random effects. Often used for hierarchical or grouped data.”
Canonical Correlation Analysis (CCA):
“Examines the relationships between two sets of variables. Discriminant Analysis:
Predicts group membership based on predictor variables (used for classification).”
Survival Analysis:
Models time-to-event data. May use extensions like Cox proportional hazards, which are not always GLMs.