Not sure if this is the right area for this post, but I wonder if anyone knows if it is justified to calculate standard deviations for "resampled" data.
I have a number of different time series plots of a certain variable from different runs (I.e. a number of y data points at certain time values). They were sampled at different times or had missing sample points, so I interpolated the data to new time points (all to the same new time points, using matlab). Can I workout the standard deviation for these new points, or can I not as they aren't "real samples". If not is there another way I can show the variance in the data?