Multivariate Analysis: Introduction and Core Concepts

Introduction to Multivariate Analysis

Chapter 1: Introduction

  • Univariate Analysis: Working with single variables or multiple variables in regression analyses.

  • Multivariate Analysis: Works with multiple variables in a multidimensional way; not the same as multiple variable analysis.

    • May be more challenging than standard statistics.

  • Outline for Topics Covered:

    • Probability review.

    • Marginal distributions.

    • Multivariate data recommendations.

    • Random factors.

    • Descriptive statistics.

  • Textbook Reference: Paragraphs 3.1, 3.2, and 4.1 are relevant to these topics.

Basic Probability Terminology

  • Random Variable: A variable whose outcomes are not determined with 100% certainty.

  • Outcome: One of the possible results.

    • Example (Single Die Roll): Six possible outcomes (1, 2, 3, 4, 5, 6); one outcome might be 44.

  • State Space: The full set of potential outcomes.

    • Example (Single Die Roll): 11 through 66.

  • Event: A subset of the state space.

  • Probability: The likelihood of an event occurring.

  • Probability Distribution: A function that maps the values of the outputs towards their probabilities.

Types of Probability Distributions

  • Discrete Distribution: For variables with discrete outcomes (e.g., 11, 22, 33, or 11 or 66).

    • Probability Mass Function (PMF): The probability distribution for discrete variables.

      • Rules for PMF:

        • The summation of probabilities for all possible states must equal 11: ∑P(xi)=1\sum P(x_i) = 1.

        • Probabilities cannot be negative: P(xi)≥0P(x_i) \ge 0.

  • Continuous Distribution: For variables with continuous, smooth outcomes (e.g., 4.54.5).

    • Probability Density Function (PDF): The probability distribution for continuous variables.

      • Rules for PDF:

        • The total area under the curve must equal 11: ∫−∞∞f(x) dx=1\int_{-\infty}^{\infty} f(x)\,dx = 1.

        • The values of the density (height of the function) cannot be negative: f(x)≥0f(x) \ge 0.

        • The height of the density function can be greater than 11 (e.g., f(x)>1f(x) > 1), as long as the total area under the curve is 11.

Describing Probability Distributions

  • Cumulative Distribution Function (CDF):

    • Formula: F(x)=P(X≤x)F(x) = P(X \le x).

    • Description: Gives the probability that a random variable XX is less than or equal to a particular value xx. It's a