Time Series Analysis for Irregularly Sampled Data

Many spectral estimation methods for irregularly sampled data tend to be heavily biased at higher frequencies or fail to produce a spectrum that is positive for all frequencies. A time series spectral estimator is introduced that applies the principles of a new automatic equidistant missing data algorithm to unevenly spaced data. This time series estimator approximates the irregular data by a number of equidistantly resampled missing data sets, with a special nearest neighbor method. Slotted nearest neighbor resampling replaces a true observation time instant by the nearest equidistant resampling time point, but only if it is within half the slot width. A smaller slot reduces the bias. Therefore, multi shift slotted nearest neighbor uses a slot width that is a fraction of the resampling time, giving equidistant data sets with slightly different starting points, shifted over the slot width. Results can be accurate at frequencies much higher than the mean data rate.