5.3 Discrete independence

Recall the definition of independence in Chapter 2.

Two events \(A\) and \(B\) are independent if

\[ \text{P}(A \mid B)=\text{P}(A) \]

By the definition of the conditional probability,

\[ \text{P}(A \mid B)=\frac{\text{P}(A \cap B)}{\text{P}(B)} \]

We have

\[ \text{P}(A \cap B) = \text{P}(A) \cdot \text{P}(B), \;\; \text{if $A$ and $B$ are independent.} \]

Two RVs \(X\) and \(Y\) are independent if

\[ p_{X \mid Y}(x \mid y)=p_X(x), \;\;\; \text{for all $x$ and $y$ with $p_Y(y)>0$.} \]

Knowing \(Y=y\) does not affect our believe on how likely \(X=x\) occurs.

By the definition of the conditional PMF,

\[ p_{X \mid Y}(x \mid y)=\frac{p_{X, Y}(x, y)}{p_Y(y)} \]

\(X\) and \(Y\) are independent if

\[ p_{X, Y}(x, y) = p_X(x) \cdot p_{Y}(y), \;\; \text{for all $x$ and $y$.} \]

  • \(X\) and \(Y\) are the rolls on two 4-sided dice
  • Are \(X\) and \(Y\) independent?
\(Y=1\) \(Y=2\) \(Y=3\) \(Y=4\) \(p_X(x)\)
\(X=1\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{4}\)
\(X=2\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{4}\)
\(X=3\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{4}\)
\(X=4\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{4}\)
\(p_Y(y)\) \(\frac{1}{4}\) \(\frac{1}{4}\) \(\frac{1}{4}\) \(\frac{1}{4}\) \(\frac{100}{100}\)

Knowing the 1st roll doesn’t give us any info about the 2nd roll.

  • \(X\) is the roll on the die #1, \(T\) is the sum of the two rolls.
  • Are \(X\) and \(T\) independent?
\(X \setminus T\) \(2\) \(3\) \(4\) \(5\) \(6\) \(7\) \(8\) \(p_X(x)\)
\(1\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(0\) \(0\) \(0\) \(\frac{1}{4}\)
\(2\) \(0\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(0\) \(0\) \(\frac{1}{4}\)
\(3\) \(0\) \(0\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(0\) \(\frac{1}{4}\)
\(4\) \(0\) \(0\) \(0\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{16}\) \(\frac{1}{4}\)
\(p_T(t)\) \(\frac{1}{16}\) \(\frac{2}{16}\) \(\frac{3}{16}\) \(\frac{4}{16}\) \(\frac{3}{16}\) \(\frac{2}{16}\) \(\frac{1}{16}\) \(1\)

Knowing the 1st roll gives us some info about the sum.