R code to analyze response time distributions as mixed models, and to cluster seqences of peaks over time as discussed in my Monitorama 2023 talk.
As initially described by this medium post and extended for Monitorama June 2023 slides here
Some basic sample data provided as a csv. Will also recognize R histogram class as input directly.
> sample <- read.csv("sample.csv")
> as.peaks(sample,plots=T,peakcount=6, normalize = T)
PeakDensity PeakBucket PeakMin PeakMax PeakMean PeakSD PeakAmplitude PeakLatency
4 0.008936170 4 3 5 3.878831 0.3263232 0.00000000 39.47890
3 0.011489362 6 5 8 6.246079 1.2272864 0.03091961 63.38428
5 0.007659574 9 8 10 8.821677 0.3827842 0.00000000 106.09506
2 0.173191489 25 20 26 24.483727 0.7152434 0.00000000 2432.67183
1 0.222553191 28 26 31 27.322500 1.0325453 0.64311539 4291.96511
