The Chi-Squared Distribution: Definition
What you'll learn: You'll discover how the chi-squared distribution is a special case of the gamma distribution, characterized by a single parameter called degrees of freedom.
What Is the Chi-Squared Distribution?
The chi-squared distribution (written as χ²) is not a completely new distribution—it's actually a specific version of the gamma distribution you just learned. While the gamma distribution has two parameters (shape α and rate β), the chi-squared distribution is defined by setting these parameters to particular values based on a single parameter called degrees of freedom, denoted by ν (the Greek letter "nu").
The Connection to Gamma
Recall that a gamma distribution has PDF involving parameters α (shape) and β (rate). The chi-squared distribution with ν degrees of freedom is defined as:
χ²(ν) = Gamma(α = ν/2, β = 1/2)
In other words:
- Shape parameter: α = ν/2
- Rate parameter: β = 1/2 (equivalently, scale parameter = 2)
The degrees of freedom ν must be a positive value (often a positive integer in practice).
Why "Degrees of Freedom"?
The term "degrees of freedom" comes from the distribution's origin: if you take ν independent standard normal random variables, square each one, and add them up, the resulting sum follows a chi-squared distribution with ν degrees of freedom. But for now, just remember that ν indexes which chi-squared distribution you're working with—higher ν means a different shape.
An Analogy
Think of the gamma distribution as a customizable template and the chi-squared as a pre-set configuration. Instead of adjusting two dials (α and β), you only turn one dial (ν), and the other settings are automatically determined.
Key Takeaway: The chi-squared distribution with ν degrees of freedom is a gamma distribution with shape α = ν/2 and rate β = 1/2, indexed by the single parameter ν.