I have been working on a Bayesian network analysis in the "gemtc" package comparing many different treatments, with little back-up from multiple articles. I've installed all of the necessary packages RJAG either. This is my data set called 'Entire':
study treatment diff std.err
1 A AA -0.3425 0.1064
2 A AB NA NA
3 B AA -0.4943 0.1086
4 B AB NA NA
5 C ADV -0.0305 0.0991
6 C AD NA NA
7 D AA -0.1054 0.1076
8 D AB NA NA
9 E AD 0.0392 0.2632
10 E AB NA NA
11 F ABV -0.01508 030685
12 F AB NA NA
13 G K -0.0325 031179
14 G AD NA NA
15 H L 030488 0.1084
15 H AB NA NA
16 I AD -0.0408 0.1078
16 I M NA NA
17 J O -0.0274 0.0967
17 J AB NA NA
Here is the necessary script:
network <- mtc.network(data.re = Entire, description = "Network")
summary(network)
nodesplit<-mtc.nodesplit(network, linearModel='random', n.adapt=10000, n.iter=100000, thin=10)
summary(nodesplit)
Once I run the mtc.nodesplit I get the following error:
Error in mtc.nodesplit(network, linearModel = "random", n.adapt = 10000, :
There are no comparisons to assess for inconsistency. See https://doi.org/10.1002/jrsm.1167 for how this is determined.
I used within the same script long time ago but it worked before. . How can I resolve this error?
I also ran the Monte Carlo simulation and plot of network, these worked without any problems.
Inconcistency test is required to assess for NMA.