White Noise Sequence

165 Views Asked by At

Would this process be a white noise sequence?

Consider the process $\{tY_t\}_{t = 1, . . . , 100}$, where $Y_t$ are independently, normally distributed with mean 0 and variance 1.

The process holds for the two first conditions which are:

  • $$ E(Y_t) = 0 $$
  • $$ Var (Y_t) = 1 $$

I am unsure about the third condition which states that:

  • $$ E(Y_t,Y_s) = 0 $$

My intuitive guess would be that this condition will still hold as the sequence is independent.

1

There are 1 best solutions below

3
On BEST ANSWER

No, the process is not white noise, because the variance is not constant, it depends on $t$. We have that $$ \operatorname{Var}[tY_t]=t^2. $$

However, the elements of this sequence are uncorrelated. Since $Y_t$ and $Y_s$ are independent, $$ \operatorname E[sX_stX_t]=st\operatorname EX_s\operatorname EX_t=0. $$ We do not need independence for the first equality, non-correlation is sufficient, but independence implies non-correlation.