How to compute a 95% confidence interval around a continuous signal?

Question

I would like to compute and display in python a 95% CI around a continuous signal (voltage values as a function of time). This signal was recorded in the brain of 16 different subjects, and lasts 1300 ms. Sampling rate was 250 Hz (so one datapoint every 4 ms). How can I proceed?

http://stackoverflow.com/questions/20033396/how-to-visualize-95-confidence-interval-in-matplotlib — dot.Py, Jun 03 '16 at 20:18
Before making useless comments, you should carefully read the question. — user1363251, Jun 03 '16 at 20:32

score 0 · Answer 1 · answered Jun 03 '16 at 20:27

Here's a pythonic example of continuous error bar plotting: https://tonysyu.github.io/plotting-error-bars.html#.V1HmMPkrJhE. The example is for plotting errors, but just replace err with stdev.

Assuming that each sample variable is normally distributed, I would calculate 2*standard deviation (~95% confidence) for each sample (every 4ms data point across all subjects). Each of these values gets stored in an array, and the array and data points can be fed into the example code.

Thanks a lot, this response is very useful. – user1363251 Jun 03 '16 at 20:33 — user1363251, Jun 03 '16 at 20:33

How to compute a 95% confidence interval around a continuous signal?

1 Answers1