设X概率密度 f(x)=(θ+1)x^θ,0

矩估计：
E(x)
=∫_(0,1) x * (θ+1)x^θ dx
=∫_(0,1) (θ+1)x^(θ+1) dx
=(θ+1)/(θ+2)*x^(θ+2) |_(0,1)
=(θ+1)/(θ+2)
令E(x)=(Σxi)/n
则θ=1/(1-(Σxi)/n) - 2
极大似然估计：
ln p(x1,x2,...,xn) = ln f(x1) + ln f(x2)...+ ln f(xn)
=n ln(θ+1) + θ Σln(xi)
对θ求导,令导数等于0得
n/(θ+1) + Σln(xi) = 0
θ = -n/Σln(xi) - 1