Cross Validated Asked on November 29, 2020
I have 100 sets of data, each of which consists of ~ 15 distributions along time, I’ve attached a figure of 1 of these sets to hopefully clear up what I mean. The plot shows the 95% confidence intervals and quartiles for each distribution, the red dot and line are actually irrelevant for this particular statistical test.
I want to determine for each of these sets whether or not between any two distributions, there is a significant difference. I originally considered the Kolmogorov Smirnov test, but for a reason I can’t remember now (which is frustrating me and I’m sure frustrating you too), that isn’t applicable. I’ve been considering the z test as outlined here: http://homework.uoregon.edu/pub/class/es202/ztest.html but i’m not sure if it is the best approach.
I also want to somehow be able to quantify the amount of sets of data across the 100 that contain significant difference, but not sure on how to do that yet (any ideas for that are most welcome).
Thank you very much
1 Asked on November 22, 2020
1 Asked on November 19, 2020 by nolw38
0 Asked on November 17, 2020 by maverick-meerkat
1 Asked on November 17, 2020 by dc_liv
0 Asked on November 14, 2020 by user819749
1 Asked on November 12, 2020 by mike-land
0 Asked on November 12, 2020 by aaryan-bhagat
0 Asked on November 9, 2020 by monica
1 Asked on November 9, 2020 by arun
1 Asked on November 8, 2020
1 Asked on November 6, 2020 by seth-c
1 Asked on November 3, 2020 by londumas
1 Asked on October 29, 2020 by thomas-lordick
0 Asked on October 26, 2020 by mattjosh
0 Asked on October 26, 2020 by michaelkyei
1 Asked on October 23, 2020 by doug-fir
0 Asked on October 20, 2020 by minstein
Get help from others!