I have a timeseries dataset of a quantity measured over the period of a week. I want to verify if the data is varying in a diurnal fashion with the help of some mathematical measure. Does any such measure exist?
closed as off-topic by David Richerby, vonbrand, András Salamon, Luke Mathieson, Juho Feb 2 '14 at 11:28
This question appears to be off-topic. The users who voted to close gave this specific reason:
- "This question does not appear to be about computer science, within the scope defined in the help center." – David Richerby, vonbrand, András Salamon, Luke Mathieson, Juho
If you want to find periodic behaviors of your data, you should use the Fourier transform, which decomposes your data into a linear combination of cyclic waveforms. If you have a guess on the actual period, as in your case, you can try something simpler. For each hour during the day, compute the mean $\mu$ and variance $\sigma^2$ of your data, and plot the intervals $[\mu-\sigma,\mu+\sigma]$ for each hour. That should give you an idea regarding your hunch. (More generally, you can try a sophisticated method like ANOVA.)
Perhaps you should ask your question over in stats.se.