PMF, expectation, variance, and the key discrete distributions
A random variable \(X\) is a function \(X: \Omega \to \mathbb{R}\) that assigns a real number to each outcome. It converts events into numbers, making probability calculations algebraic.
\(p_X(x) = \mathbb{P}(X = x) = \mathbb{P}(\{\omega \in \Omega : X(\omega) = x\})\)
Valid PMFs satisfy: \(p_X(x) \geq 0\) and \(\sum_x p_X(x) = 1\).
The expected value is the probability-weighted average. For a function \(g(X)\): \(\mathbb{E}[g(X)] = \sum_x g(x)\, p_X(x)\).
\(\mathbb{E}[aX + b] = a\,\mathbb{E}[X] + b\). Holds even for dependent random variables.
\(X \sim \text{Ber}(p)\): single trial, success with probability \(p\).
\[p_X(k) = \begin{cases} p & k=1 \\ 1-p & k=0 \end{cases}, \quad \mathbb{E}[X]=p, \quad \text{Var}(X)=p(1-p)\]\(X \sim \text{Bin}(n,p)\): number of successes in \(n\) independent Bernoulli(\(p\)) trials.
\[p_X(k) = \binom{n}{k}p^k(1-p)^{n-k}, \quad k=0,1,\ldots,n\] \[\mathbb{E}[X]=np, \quad \text{Var}(X)=np(1-p)\]\(X \sim \text{Geo}(p)\): number of failures before first success. \(X \in \{0, 1, 2, \ldots\}\).
\[p_X(k) = (1-p)^k p, \quad \mathbb{E}[X] = \frac{1-p}{p}, \quad \text{Var}(X) = \frac{1-p}{p^2}\]Memorylessness: \(\mathbb{P}(X > s+t \mid X > s) = \mathbb{P}(X > t)\). Past failures carry no information about future trials.
\(X \sim \text{Unif}(a, b)\): equally likely values \(\{a, a+1, \ldots, b\}\).
\[\mathbb{E}[X] = \frac{a+b}{2}, \quad \text{Var}(X) = \frac{(b-a)(b-a+2)}{12}\]The conditional PMF of \(X\) given event \(A\): \(p_{X|A}(x) = \mathbb{P}(X=x \mid A)\).
For a partition \(A_1, \ldots, A_n\) of \(\Omega\):
\[\mathbb{E}[X] = \sum_i \mathbb{P}(A_i)\,\mathbb{E}[X \mid A_i]\]
The expected value is a weighted average of conditional expectations — one of the most useful tools for computing expectations of complex processes.
| Distribution | PMF \(p_X(k)\) | \(\mathbb{E}[X]\) | \(\text{Var}(X)\) |
|---|---|---|---|
| Bernoulli(\(p\)) | \(p^k(1-p)^{1-k}\) | \(p\) | \(p(1-p)\) |
| Binomial(\(n,p\)) | \(\binom{n}{k}p^k(1-p)^{n-k}\) | \(np\) | \(np(1-p)\) |
| Geometric(\(p\)) | \((1-p)^k p\) | \((1-p)/p\) | \((1-p)/p^2\) |
| Discrete Uniform(\(a,b\)) | \(1/(b-a+1)\) | \((a+b)/2\) | \((b-a)(b-a+2)/12\) |