Removings spikes in a data set

Hi,
I am wondering is there any way to remove random spikes in the data.?
If someone can help me I can send out a sample of my data set.
Your help is highly regarded
Thanks
Thisara Start a New Thread

I suppose one could:

1. selectively throw out all data points greater than a particular
absolute value
2. selectively throw out all points that differ from the mean by a
certain amount (will not work if your mean is time varying)
3. filter the data to smooth it (though this will impact the non spikey
points as well).

It is important in my opinion to be really sure that you want to "remove the
spikes". You might be removing data that is crucial to explain some real
phenomenon in your data set.

thanks
On Tue, Nov 9, 2010 at 6:55 PM, thisara w wrote:

> Hi,
> I am wondering is there any way to remove random spikes in the data.?
> If someone can help me I can send out a sample of my data set.
> Your help is highly regarded
> Thanks
> Thisara Start a New Thread