Expectation of a transformation of normal random variable X does not depend on the mean of X

38 Views Asked by At

I want to know if there exist a function $g$ and $\mu \ne \mu’$ such that $$ \int g(x)\phi(x;\mu)\,dx = \int g(x)\phi(x;\mu’)\, dx$$ Where $\phi(x;\mu)$ is the pdf of normal random variable with mean $\mu$ and unit variance.

All I can find is that if $g$ is symmetric — $g(x) = g(-x)$, then the above condition holds.

Could there be other functions that satisfy the above equation?

1

There are 1 best solutions below

1
On

There is such a function $g$. Take $g(x)=x-\frac 1{\mu+\mu'} x^{2}$. The case $\mu+\mu'=0$ is left to you.