Convergence in distribution

Summary

  • z1,z2, is a sequence of random variables where zn has cumulative distribution function Fn(z) for n=1,2, and z is a random variable with cumulative distribution function F(z) .
  • We say that the sequence z1,z2, converges in distribution to a random variable z if

Fn(z)F(z)

  • as n for all zR at which F is continuous.

Central limit theorem

  • (Lindeberg–Lévy CLT). x1,x2,...,xn is an IID random sample, E(xi)=μ and Var(xi)=σ2 . Then

n(¯xnμ)N(0,σ2)

  • or

n(¯xnμ)σN(0,1)

  • For n large, ¯xn will be approximately N(μ,σ2/n) .