- #1
Matteo_
- 1
- 0
Hi everybody!
I have a random iid sample Xi, i=1, ..., n
The empirical cdf of the sample at poin s is
[tex]\hat{F}\left(s\right)=n^{-1}\sum_{i=1}^{n}{\textbf{1}_{\left(-\infty, s\right)}\left(x_{i}\right)}[/tex]
Clearly [tex]\hat{F}\left(s\right)[/tex] is binomially distributed with parameters n and p=F(t) (true cdf).
Now I need to find the covariance between [tex]\hat{F}\left(s\right)[/tex] and [tex]\hat{F}\left(t\right)[/tex] for s<t.
I know that the result is [tex]n^{-1}F\left(s\right)\left(1-F\left(t\right)\right)[/tex]
Any help is very appreciated...
Thanks!
I have a random iid sample Xi, i=1, ..., n
The empirical cdf of the sample at poin s is
[tex]\hat{F}\left(s\right)=n^{-1}\sum_{i=1}^{n}{\textbf{1}_{\left(-\infty, s\right)}\left(x_{i}\right)}[/tex]
Clearly [tex]\hat{F}\left(s\right)[/tex] is binomially distributed with parameters n and p=F(t) (true cdf).
Now I need to find the covariance between [tex]\hat{F}\left(s\right)[/tex] and [tex]\hat{F}\left(t\right)[/tex] for s<t.
I know that the result is [tex]n^{-1}F\left(s\right)\left(1-F\left(t\right)\right)[/tex]
Any help is very appreciated...
Thanks!