Feature request: 95% confidence interval averages

[pklose]

Hello John and Forum,

often I use the average function in REW. It would be beneficial to my workflow (and maybe to others) to have an additional "95% confidence average" option in the average calculation beside dB average, RMS average and vector average. Thanks!

 

[John Mulcahy]

Not sure what you mean by that. The 95% confidence interval would be a property of each frequency value in the average based on the values of the traces being averaged at that frequency. Given the number of traces being averaged is usually small I'm not sure it would be particularly meaningful.

 

[pklose]

Hi John, you are right. Usually the number of traces used for averaging is small, so the 95% average would not be really useful there. But I am seeing often in-room measurements and in-car measurements with a high number traces (+16 e.g. with the Umik-X). And a not so-optimal-position outlier could be eliminated with this feature. It would be a good addition to the existing min/max grey tubes.