I want to estimate 95% C.I. for RMSE(Root mean square error).
However, i have a difficulty in estimating C.I. due to RMSE equation
(there is a root in RMSE equation)
I want to estimate 95% C.I. for RMSE(Root mean square error).
However, i have a difficulty in estimating C.I. due to RMSE equation
(there is a root in RMSE equation)
You can use the same approach that is common to get confidence intervals for standard deviation: we get confidence intervals for variance and then just take the square root of the extremes of the interval.
Here you can compute a CI for mean square error and take the square root of the values you get. Since square root is a monotonic growing function, probabilities inside and outside of the interval remain the same.