Will more data rather than more samples improve spectroscopic analysis?

It is well known that the noise on a spectrometric signal will decrease with averaging over time. With contemporary capability for data collection and storage, we can retain and access more information about a signal train than just its average. During the same sampling time, we can record multiple versions of the signal averaged over shorter, equal periods. This is, then, the set of signals over submultiples of the total collection time. With a sufficiently large set of submultiples, the distribution of the signal's fluctuations over the submultiple periods of the data stream can be acquired at each wavelength. We have previously shown that the extreme values of the fluctuation of the signals are usually not balanced (equal magnitudes, equal probabilities) on either side of the mean or median without an inconveniently long measurement time; the data is almost inevitably biased away from the mean indicating benefits from using the median. Here, we explore the use of submultiple data collection to improve multivariate curve resolution – alternating least squares methods to separate the infrared spectra of buffer and protein, objectively, from protein solution spectra.