Conditional variance of Y given X when y is a continuous function of x

291 Views Asked by At

Technically the problem term would be $E[Y^2|X]$

For simplicity, let us for a minute assume that $y = a + bx$

Is it mathematically correct to write: $E[Y^2|X] = E[(a+bx)^2]$, and if so why?

And how would this apply to the direct formula of variance? Can we formulate

$$V[Y|X] = E[(Y - E[Y|X])^2 \ | \ X] = E[((a+bx)-E[Y|X])^2 ]$$

Also does anyone have a link where I can read more about this?

1

There are 1 best solutions below

0
On BEST ANSWER

A conditional expectation is a function of the conditioning variable. Here, that is $X$.

For any function, $g$, the conditional expectation $\mathsf E(g(X)\mid X)$ will equal $g(X)$.

[ When given $X$, then you should surely expect $g(X)$ to be $g(X)$. ]

So when $Y$ is $a+bX$, then: $$\mathsf E(Y^2\mid X)=(a+bX)^2$$


Likewise for the conditional variance.$$\begin{align}\mathsf{Var}(Y\mid X) &= \mathsf E((Y-\mathsf E(Y\mid X))^2\mid X) \\ &=\mathsf E(((a+bX)-\mathsf E(a+bX\mid X))^2\mid X)\\&=0\end{align}$$

Note: You should anticipate this result. Since for any given value of $X$, we have that $Y$ is invariably $a+bX$.