Scribd
Upload
15 views2 pages

Handwritten Style Multivariate Notes

The document provides an overview of various multivariate techniques including ANOVA, MANOVA, Discriminant Analysis, PCA, Factor Analysis, and Cluster Analysis, each with examples and key concepts. It also includes a quick revision formula summarizing the purpose of each technique and lists common assumptions for their application. The techniques are essential for analyzing multiple variables and understanding relationships within data.

Uploaded by

sohaib7864u
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
15 views2 pages

Handwritten Style Multivariate Notes

The document provides an overview of various multivariate techniques including ANOVA, MANOVA, Discriminant Analysis, PCA, Factor Analysis, and Cluster Analysis, each with examples and key concepts. It also includes a quick revision formula summarizing the purpose of each technique and lists common assumptions for their application. The techniques are essential for analyzing multiple variables and understanding relationships within data.

Uploaded by

sohaib7864u
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

✍ Multivariate Techniques – Exam Friendly Notes

1. ANOVA

■ Compare means of 2 or more groups using ONE dependent variable.


Example: Compare average marks of Science, Arts and Commerce students.
Remember: ANOVA = Difference in means.

2. MANOVA

■ Extension of ANOVA with MULTIPLE dependent variables.


Example: Compare students on Marks + Attendance + Confidence together.
Remember: MANOVA = Multiple outcomes.

3. Discriminant Analysis

■ Classifies observations into existing groups.


Example: Predict whether a student belongs to Pass or Fail group.
LDA: Equal covariance matrices.
QDA: Different covariance matrices.

4. PCA (Principal Component Analysis)

■ Reduces many variables into a few components.


Example: 20 survey questions → 3 main components.
Remember: PCA = Data Reduction.

5. Factor Analysis

■ Finds hidden factors behind correlations.


Example: IQ, reasoning and memory may represent one hidden factor: Intelligence.
Remember: FA = Hidden Causes.

6. Cluster Analysis

■ Creates groups automatically (no labels required).


Example: Customers grouped into Premium, Regular and Budget buyers.
K-Means: Fixed number of clusters.
Hierarchical: Tree (Dendrogram) based clustering.

Quick Revision Formula


ANOVA → Difference
MANOVA → Multi-Difference
Discriminant → Classification
PCA → Reduction
Factor Analysis → Hidden Factors
Cluster → Grouping

Common Assumptions

✓ Independence
✓ Normality
✓ No extreme outliers
✓ Adequate sample size
✓ Linear relationships
✓ Low multicollinearity

You might also like