Statistical independence

From Maths
Jump to: navigation, search
Stub grade: A*
This page is a stub
This page is a stub, so it contains little or minimal information and is on a to-do list for being expanded.The message provided is:
I want to phrase this intuitively, put that statement formally, then show that's equiv to claim 1 below.

I also want to explore how suppose P[A|B]=P[A] is all we know, what can we say about independence then? Alec (talk) 13:47, 17 October 2017 (UTC)

Main reference: Template:RMSADAR - Mathematical Statistics and Data Analysis - 2nd edition - John A. Rice Alec (talk) 13:47, 17 October 2017 (UTC)

\newcommand{\P}[1]{\mathbb{P}\left[#1\right]}


Definition

Let (S,\Omega,\mathbb{P}) be a probability space and let A,B\in\Omega be events. We say "A and B are statistically independent if:

  • The probability of A is the same as the probability of A given B has occurred, and,
    the probability of B is the same as the probability of B given A has occurred.
    • We may write this symbolically (conditional probability) as:
      1. \mathbb{P}[A]\eq\mathbb{P}[A\vert B]
      2. \mathbb{P}[B]\eq\mathbb{P}[B\vert A]

Formally:

  • For A,B\in\Omega, A and B are independent if \big[\big(\mathbb{P}[A]\eq\mathbb{P}[A\vert B]\big)\wedge \big(\mathbb{P}[B]\eq\mathbb{P}[B\vert A]\big)\big][Note 1]
    • Claim 1: A and B are statistically independent \iff\big[\mathbb{P}[A\cap B]\eq\mathbb{P}[A]\cdot\mathbb{P}[B]\big]
      • Formally: \forall A,B\in\Omega\left[\big(\P{A\vert B}\eq\P{A}\wedge\P{B\vert A}\eq\P{B}\big)\iff\big(\P{A\cap B}\eq\P{A}\cdot\P{B}\big)\right]

Proof of claims

Claim 1:

Claim: \forall A,B\in\Omega\left[\big(\P{A\vert B}\eq\P{A}\wedge\P{B\vert A}\eq\P{B}\big)\iff\big(\P{A\cap B}\eq\P{A}\cdot\P{B}\big)\right]

Notes

  1. Jump up See Definitions and iff, as in fact they are independent \iff this

References