Re: Jointly Gaussian Random Variables
- From: "sarwate@xxxxxxxxxxxxxxxx" <dvsarwate@xxxxxxxxx>
- Date: Sun, 2 Mar 2008 10:12:54 -0800 (PST)
On Feb 29, 6:24 pm, "Manolis C. Tsakiris" <el01...@xxxxxxxxxxxx>
wrote:
Hello,
let x and y be zero-mean, jointly Gaussian random variables and let E[] be
the expectation value operator. Then is it true that
E[(x^3)y] = E[xy] * E[x^2] ?
Hint: E[X^3Y] = E[E[X^3|Y]}. Conditioned on the value of Y, X is a
Gaussian random variable (with nonzero mean m that depends on the
value of Y). The value of E[X^3|Y] is m^3 + 3ms^2 where s^2 is the
conditional variance of X given the value of Y. The result is a
function
of Y Then, compute the expectation of this function of Y and see if
your identity holds or not.
Hope this helps
Dilip Sarwate
.
- Follow-Ups:
- Re: Jointly Gaussian Random Variables
- From: Manolis C. Tsakiris
- Re: Jointly Gaussian Random Variables
- References:
- Jointly Gaussian Random Variables
- From: Manolis C. Tsakiris
- Jointly Gaussian Random Variables
- Prev by Date: Re: how to derive IFFT of boxcar filter
- Next by Date: Re: Low pass filter at half Nyquist
- Previous by thread: Re: Jointly Gaussian Random Variables
- Next by thread: Re: Jointly Gaussian Random Variables
- Index(es):
Relevant Pages
|