id data metric ref test type ref.m1 ref.m2 ref.m3 ref.m4 ref.c1 ref.c2 ref.c3 test.nexp test.m1 test.m2 test.m3 test.m4 test.c1 test.c2 test.c3 t2.1 t2.2 t2.3 t2.4 t2.5 t2.6 t2.7 t2.8 t2.9 t3.1 t3.2 t3.3 t3.4 t3.5 t3.6 t3.7 t3.8 t3.9 prop outlier bioq
44 d1a tmax 0.7 3 opt -0.9857393 -1.92927 NA NA 3.445674 NA NA 2 -0.9830524 -28.50164 NA NA 3.374213 NA NA 0 0.1 0.2 2.1 3.6 5.9 10.4 11.5 24 0 0.1 0.2 1.4 2.4 3.8 5.9 9.9 24 4.285714 TRUE FALSE
source(paste(git.dir, reponame, "outliers/outlier_functions.R", sep = "/"))
ref.par <- c(-0.9857393, -1.92927, 3.445674)
fit.par <- c(-0.9830524, -28.50164, 3.374213)
t2.int <- c(0, 0.1, 0.2, 2.1, 3.6, 5.9, 10.4, 11.5, 24)
t3.int <- c(0, 0.1, 0.2, 1.4, 2.4, 3.8, 5.9, 9.9, 24)
conc.data <- matrix(c(pred.d1a(time.samp, ref.par), pred.d1a(time.samp, fit.par)), ncol = 2)
plot.sumexp.overlay(conc.data, time.samp, 2, -1)
Check intervals, they match well to the fitted parameters, but not the reference.
plot.sumexp.facet(conc.data, time.samp, 2, t2.int)
plot.sumexp.facet(conc.data, time.samp, 2, t3.int)
source(paste(git.dir, reponame, "fn_diag/fix_functions.R", sep = "/"))
cdf <- data.frame(time.samp, conc.data[,1])[(1:9)*30-29,]
chisq.sumexp(optim.sumexp(cdf, oral = T))$par
Unsure why the popsize run didn't work, considering that running the script manually says it should have worked! EDIT (17/29/6): Error in broad_analysis not accounting for number of iterations