[?] Calculate confidence interval for an expected value in linear model?

Questions
,
1

Hi everyone

I’m sure there is a really simple solution for my problem but I just can’t figure it out.

After using LinearModelFit to estimate a mode (lm)l, I calculated an expected point estimate using some fixed values (lm[1,2,3]). I’d now like to calculate a confidence interval for my point estimate. What would be the way to do it?

Thanks

2

An explicit example would be helpful here. That should include some indication of the desired result.

3

Hi Daniel

Thanks for the reply. So after estimating a model, I get a value of lets say 13 - I now want the 95% CI for this value, so what I’m searching for is basically a confidence interval for a point estimate. Does that make sense?

Thanks

4

Luka,

I think what you want is in the documentation here: LinearModelFit

Search for “MeanPredictionConfidenceIntervalTable” in the examples and read the next few sections-- it shows how to get a table of confidence bands and even confidence functions for a given ConfidenceLevel with MeanPredictionBands.

As Daniel posted, to be of any more help, you would be best to give an actual example with some data and what specifically you are trying to do.

Regards,

Neil