<< probability_theory

Continuous Distribution

1. Normal Distribution

For the random variable with probability density function p(x)=12πσexp{(xμ)22σ2} where μ,σ>0 are parameters, we define X complies the normal distribution XN(μ,σ2).

Regularity Checking.

Let x=xμσ, F(+)=p(x)dx=12πσexp{(xμ)22σ2}dx=φ(x)=12πex22 Let x=2t, F(+)=

Specially, when μ=0,σ=1, the normal distribution is called standard normal distribution, denoted as XN(0,1). The probability density function of standard normal distribution is denoted as φ(x)=12πex22 and its cumulative distribution function denoted as Φ(x). Notice that φ(x) is even function hence Φ(x)+Φ(x)=1.

Any normal distribution X(μ,σ2) can be converted to the standard form by linear transformation: Y=XμσN(0,1)

Proof. The cumulative distribution function of constructed random variable Y is FY(y)=P(Xμσy)=P(Xσy+μ)=0σy+μ12πσexp{(xμ)22σ2} Let x=σy+μ, we have FY(y)=0y12πσey22 and the probability density function pY(y)=dFY(y)dy=d(0y12πσey22)dy=12πσey22N(0,1)

The most simple form of normal distribution is

2. Uniform Distribution

The uniform distribution is random variable with probability density function p(x)={1baaxb0,x<a,x>b denoted as XU(a,b).

3.