unequal resampling

Share icon

Mon, 01/07/2019 - 10:18 am

Hi Everyone,

I have two climate time series that I wish to compare (wave 1, wave 2) and ultimately perform a correlation on the represented variables. The data in both series have unequal time increments (i.e. the spacing between each data point is different), and one series (wave 1) has a much higher resolution than the other (wave 2). The lower-resolution wave 2 is a continuous time series and each point integrates a variable amount of time; wave 1 is a discrete time series, but its resolution is high enough to be treated as continuous (for the purpose of my study). I would like to resample wave 1 to the time scale of wave 2. Does anyone know a way of performing the Resample command using unequal decimation (e.g. by somehow using the time scale of wave 2)? I have tried Interpolate but that is no good because it does not accommodate the fact that each lower-resolution data of wave 2 integrates relatively large (and unequal!) amounts of time compared to wave 1. If anyone has any bright ideas, I would love to hear from you.

 

Regards

RND

Perhaps you could integrate (with respect to time) the response data in wave1 (i.e. calculate a cumulative response curve), Resample or interpolate this to the time scale of wave2, and finally differentiate (with respect to time) to restore the data back to being response, but as if it were sampled at wave2's time intervals.

Hope this helps,

Kurt

January 7, 2019 at 10:55 pm - Permalink