Statistical Inference-Univariate Random Variables

1.1 Discrete Random Variables
A random variable is discrete if it takes values in some countable set,e.g.,{0,1,2,…}.

The mass function(pdf) $f_X(x)$ of a discrete random variable $X$ is

$$f_X(x)=P(X=x)$$

It has three properties:

(1)$f(x)\ge 0$(Positivity)

(2)${\sum }_{x}f(x)=1$(Unitarity)

(3)$P(X\in A)={\sum }_{x\in A}f(x)$(Additivity)

Expected value(mean) denoted interchangeably by $E(X)$ or ${\mu }_X$,is defined as
$${\mu }_X=E(X)=\sum _{x}xf(x)$$

Theorem 1.1

If $a,b,a_1$ and $a_2$ are constants, $h(x),h_1(x)$ and $h_2(x)$ are functions,then

$$E[a_1{h}_1(X)+a_2{h}_2(X)]=a_{1}E[h_1(X)]+a_{2}E[h_2(X)]$$

The variance(square of the standard deviation) is defined as

$${\sigma }_X^2=Var(X)=E[(X-{\mu }_X{)}^2]$$

For a discrete random variable,

$$E[(X-{\mu }_X{)}^2]=\sum _x(x-{\mu }_X{)}^{2}f(x)$$

Theorem 1.2

$$Var(X)=E(X^2)-{\mu }_X^2$$