I have data regarding the peak flood in a watershed over 40 events (Non-uniform distribution between seasons). I wish to explore the significant difference in peak flood values between seasons.
Obviously there is a small difference between seasons, but how do I prove that?
Any solution would be welcome (Do you have any suggestions for R packages?).
Multiple comparisons for non-normal distribution, one solution might be a Kruskall-Wallis with a Dunn's post-hoc and bonferroni correction? Maybe seasons as categorical factor? For 'R' language, the PMCMR package is straight forward: https://cran.r-project.org/web/packages/PMCMR/PMCMR.pdf
Do you have time-series data? Another option could be auto-correlation within a year or cross-correlation between years.
EflowStats is a useful tool for calculating ecological flow statistics for flow events. https://github.com/USGS-R/EflowStats
Thank you David,
I have figured it out that "One-way ANOVA" should be used (Going through google and if I understand correctly). Your first suggestion is more relevant (I will work on that).
"Do you have time-series data?", NO, I have just peak flow value in each flood event.
Also the "EflowStats" package in in line with my interests. I will try that to other daily data set.
Be sure to check the assumptions of the ANOVA before committing. Otherwise, as David suggests, you will need to do the non-parametric version (KW) test.
The assumptions of one way ANOVA are
Each sample is an independent random sample.
The distribution of the response variable follows a normal distribution.
The population variances are equal across responses for the group levels.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS