Hello,
I have two time-series, A and B; what test can I apply to check if they differ?
I could compare the values between A and B at specific time points; for instance, is there a difference between the measures for A and B at t=24 hours?
But, is there a better way to compare time points?
Thank you
Ways to compare time series are:
There are several ways to compare time series. Some of them are:
Thank you. These go a bit beyond my stat understanding. The easiest approach would be the Pearson correlation, but I understand it is not really good for time series. Is there a time-specific comparison test (for instance, to compare the points at t=24 or 48 h), or is a t-test enough? The other tests look to me as advanced regression methods wich go beyond my original question (tell if two series are different).
I am afraid that the suggestions given by Chuck A Arize do not answer the question. You would like a kind of t-test but the problem is that the two time series are not random samples. If the series are stationary, it can probably be done but taking the autocorrelation into account but I don't know how. If the time series are not stationary, there is little hope. Good luck
Ibragimov and Muller (2010) proposed a general procedure for robust inference about a scalar parameter of interest when the series exhibits potential heterogeneity and correlation in a largely unknown way.
See:
Ibragimov R, Muller UK. t-Statistic Based Correlation and Heterogeneity Robust Inference. Journal of Business and Economic Statistics. 2010; 28:453–468. https://doi.org/10.1198/jbes.2009.08046
Also:
Ibragimov R, Muller UK. Inference with Few Heterogeneous Clusters. Review of Economics and Statistics. 2016; 98:83–96. https://doi.org/10.1162/REST_a_00545
There is R implementation. See the package ‘clusterSEs' in R.
- Badges
- Science topic
- Similar topics
- Geoscience
- Atmospheric Sciences
- Climate Science
- Climate Change