Let's say we have two distribution f(x)β and g(x).βNow the problem is let's say you want to estimate the expectation of g(x)βbut you can't sample from it, hence you won't be able to use sample mean to estimate the expectation. So what we do here is, we use samples from f(x)βto estimate the expectation of g(x)β.
So we know we can estimate the expectation of f(x)β using sample mean by:
Efβ[x]βn1βi=1βnβxiβ,Β xiββΌf(x) βNow to estimate Egβ[x] we go trick
Egβ[x]=βxg(x)=βxf(x)g(x)βf(x)=Efβ[xf(x)g(x)β]βn1βi=1βnβxiβf(xiβ)g(xiβ)β,Β xiββΌf(x) Importance Sampling to estimate probability using Monte Carlo Method