Good morning,
I hope this message finds you well.
I am writing because I have been trying to run a Bayesian network
meta-analysis for 2 weeks now using the pcnetmeta package and have not been
able to do it. I have looked for help online and could not find what my
problem is.
My code is:
#### Bayesian Network Meta-analysis with follow up times
LR <- read.csv("C:/Users/Galanova/Box Sync/IHME/EF meta/R
analysis/diabetes.csv", row.names=NULL, quote="")
View(LR)
nma.ab.followup(s.id = s.id, t.id = t.id, event.n = event.n, total.n total.n,
followup = follow.up,
data=LR, param = c("lograte", "rank.prob"),
model = "het_eqcor", prior.type="invgamma", a =
0.001, b = 0.001,
c = 10, higher.better = FALSE, digits = 4, n.adapt = 5000,
n.iter = 100000, n.burnin = floor(n.iter/2), n.chains = 3,
n.thin = max(1, floor((n.iter - n.burnin)/100000)),
conv.diag = FALSE, trace = "lograte", dic = FALSE,
postdens = FALSE, mcmc.samples = FALSE)
My data is:
'data.frame': 128 obs. of 11 variables:
$ Author : Factor w/ 124 levels "Ali (2011)-1000 mg",..: 70 69 39
38 102
101 104 103 106 105 ...
$ s.id : int 1 1 2 2 3 3 4 4 5 5 ...
$ t.id : int 1 0 1 0 1 0 1 0 1 0 ...
$ trtname : Factor w/ 5 levels "ALT","Control",..: 3 2 3 2
3 2 3 2 3 2 ...
$ event.n : num NA NA 34 55 8 18 35 55 15 36 ...
$ total.n : int 108 80 143 142 52 54 300 321 103 110 ...
$ follow.up: num 1 1 3.1 3.1 3 3 3 3 3 3 ...
$ X1000.py : num NA NA 80 132 60 127 42 58 27 51 ...
$ percent : num NA NA 24 39 18 38 12 17 15 8 ...
$ rr : num NA NA 0.58 NA 0.42 NA 0.56 NA 0.53 NA ...
$ rr.se : num NA NA 0.55 NA -0.06 NA 0.14 NA -0.0325 NA ...
When I run it, I get this message:
Error in jags.model(file = textConnection(modelstring), data = data.jags,
:
Error in node y[17]
Observed node inconsistent with unobserved parents at initialization.
Try setting appropriate initial values.
I have tried using different initial values and that does not work either. I
am a beginner in R and would really appreciate some help on this.
Thank you for taking the time to read this message.
Karla
[[alternative HTML version deleted]]