I am looking at the errors of my model, i.e. difference between predicted outputs and actual values. Finding a mean and standard deviation, I found that many values (sometimes more than 50%) are outside of 95% confidence interval. Does this mean that my error is no normally distributed? Is that even possible? Shouldn't 95% CI mean 95% of my values be in this range?
Best Answer
It doesn't mean that there isn't a problem, but you are comparing apples with oranges. The confidence interval is for the mean — not the population. With a huge amount of data, the confidence interval for the mean will be very narrow because you can estimate the mean very accurately — but almost all the data values will be outside that confidence interval.
Put another way, the confidence limits are about $pm 2sigma/sqrt{n}$, while the 95% limits for a normal population are about $pm 2sigma$, without dividing by $sqrt{n}$.
Similar Posts:
- Solved – Does high number of values outside of 95% Confidence Interval imply non-normality
- Solved – Does high number of values outside of 95% Confidence Interval imply non-normality
- Solved – It is possible to find point estimate of population mean and population variance when confidence interval of population mean is given
- Solved – Confidence Interval of a Lognormal Random Variable
- Solved – Confidence Interval of a Lognormal Random Variable