Let X1,…,XnX_1,\ldots,X_n be nn random variables defined on a common probability space (Ω,F,P)\left(\Omega,\mathcal{F},\mathbb{P}\right) with a common state space SS and PX\mathbb{P}_X be a probability distribution on S.S. The random vector X‾=(X1,…,Xn)\underline{X}=\left(X_1,\ldots,X_n\right) is a random sample of size nn from PX\mathbb{P}_X if and only if X1,…,XnX_1,\ldots,X_n are iid with distribution PX.\mathbb{P}_X.

Let X‾\underline{X} be a random sample from PX.\mathbb{P}_X. The distribution PX\mathbb{P}_X is the population of the random sample.

Let X‾\underline{X} be a random sample of size n.n. An observed random sample is a vector x‾=(x1,…,xn)\underline{x}=\left(x_1,\ldots,x_n\right) of realised values of X‾.\underline{X}. A set of data is modelled as an observed random sample.

Let X‾\underline{X} be a random sample from the population PX.\mathbb{P}_X. The population is parametric if and only if PX\mathbb{P}_X is assumed to belong to a parametric family {PX; θ:θ∈Θ}\left\{\mathbb{P}_{X;\,\theta}:\theta\in\Theta\right\} indexed by a scalar or vector parameter θ\theta with parameter space Θ.\Theta. The population is written PX; θ\mathbb{P}_{X;\,\theta} to show that its pmf or pdf depends on θ.\theta.

Let X‾\underline{X} be a random sample of size nn with state space SS and t:Sn→Rkt:S^n\rightarrow\mathbb{R}^k be a function for some k≥1k\geq 1 that does not depend on any unknown parameter. A statistic is the random variable or random vector T=t(X‾).T=t\left(\underline{X}\right).

Let TT be a statistic. The sampling distribution of TT is the probability distribution of T.T.

Let X‾\underline{X} be a random sample of size n.n. The sample mean of X‾\underline{X} is the statistic given by the arithmetic mean of the random sample.

X‾n:=1n∑i=1nXi\overline{X}_n:=\frac{1}{n}\sum_{i=1}^n X_i

Let X‾\underline{X} be a random sample of size n≥2.n\geq 2. The sample variance of X‾\underline{X} is the statistic given by the sum of the squared deviations from the sample mean divided by n−1.n-1.

Sn2:=1n−1∑i=1n(Xi−X‾n)2S_n^2:=\frac{1}{n-1}\sum_{i=1}^n \left(X_i-\overline{X}_n\right)^2

Let X‾\underline{X} be a random sample of size n.n. The order statistics X(1)≤X(2)≤…≤X(n)X_{(1)}\leq X_{(2)}\leq\ldots\leq X_{(n)} of X1,…,XnX_1,\ldots,X_n are statistics of the random sample, and the rthr^\mathrm{th} order statistic is X(r).X_{(r)}.

Let X‾\underline{X} be a random sample of size n.n. The sample minimum and sample maximum of X‾\underline{X} are the smallest and largest order statistics.

X(1)=min⁡i∈{1,…,n}Xi,X(n)=max⁡i∈{1,…,n}XiX_{(1)}=\min_{i\in\left\{1,\ldots,n\right\}} X_i,\qquad X_{(n)}=\max_{i\in\left\{1,\ldots,n\right\}} X_i

If the population is continuous then ties occur with probability 0,0, so X(1)<X(2)<…<X(n)X_{(1)}<X_{(2)}<\ldots<X_{(n)} with probability 1.1.

Let x‾\underline{x} be an observed random sample of size nn with order statistics x(1)≤x(2)≤…≤x(n).x_{(1)}\leq x_{(2)}\leq\ldots\leq x_{(n)}. The sample median of x‾\underline{x} is the middle order statistic when nn is odd and the average of the two middle order statistics when nn is even.

m:={x(n+12)if n is odd12(x(n2)+x(n2+1))if n is evenm:=\begin{cases} x_{\left(\frac{n+1}{2}\right)} & \mathrm{if}\,n\,\mathrm{is}\,\mathrm{odd}\\ \frac{1}{2}\left(x_{\left(\frac{n}{2}\right)}+x_{\left(\frac{n}{2}+1\right)}\right) & \mathrm{if}\,n\,\mathrm{is}\,\mathrm{even} \end{cases}