Saving numerical kernel density

[glu]

Is there a possibility to save numerical results of a kernel density estimation in EViews 6 like in EViews 5 and earlier?

[EViews Gareth]

Proc->make distribution data

[glu]

Thanks a lot!

[mintyorbit]

Sorry to bring this thread up again after all this time, but I am wondering whether there is a way of finding the underlying data for two charts which are superimposed on one another. For example, I am currently trying to draw the kernel distribution for a set of data, and then draw the corresponding normal distribution on top of the kernel so that it is easy to compare them. I can do this easily in EViews by going to Graph -> Distribution -> Options and then adding the two distributions together. But once this is drawn in EViews, how can we extract the underlying data so that the graph can be drawn in Excel?

If anyone knows the answer to this I would be very grateful to them for any advice.

[EViews Gareth]

Proc->make distribution data

[EViews Glenn]

To add to Gareth's comment, you'll need to do this for each layer.

[mintyorbit]

Thank you both for your replies. I had indeed tried this, as this is what you had recommended for the previous participant in this thread. The trouble is that when you do this separately for each series, the normal distribution is no longer calibrated to fit the emprical data.

If there is no way around this I shall simply do this by hand. Thank you again for your advice in this matter.

[EViews Glenn]

I guess that's right. Shouldn't be very hard. All you have to do is to take the evaluation points used in the kernel, standardize using the moments of the data, then run through the @dnorm.