Thread Subject: CDF of a signal

I need to calculate the cdf of a signal which is from a industry. I tried using the CDF command but it does not work as the signal data does not follow any certain distribution. I think maximum likelihood estimation would work, but not sure how to use that :(

Any suggestion/advice?

Thanks.

On Nov 15, 9:00 pm, "naseeb " <adnan....@gmail.com> wrote:
> I need to calculate the cdf of a signal which is from a industry. I tried using the CDF command but it does not work as the signal data does not follow any certain distribution. I think maximum likelihood estimation would work, but not sure how to use that :(
>
> Any suggestion/advice?
>
> Thanks.

I'm not sure what "a signal which is from a industry" means, but if
you mean it is real data, then you probably need the empirical cdf
which you can calculate using:
help hist - to calculate the pdf
help cumsum - to integrate the pdf into its cdf

naseeb wrote:
> I need to calculate the cdf of a signal which is from a industry. I tried using the CDF command but it does not work as the signal data does not follow any certain distribution. I think maximum likelihood estimation would work, but not sure how to use that :(

Naseeb, since you mention the CDF function, it seems as if you have access to the Statistics Toolbox. It sounds like you want to make some kind of non-parameteric estimate of a CDF based on your data. I think what you're looking for is either the ECDF function, or the KSDENSITY function, the latter with the 'Function' parameter set to 'CDF.

Hopsing