I am new using INLA. This is an exercise from Faraway's book. When I copy and paste the example code in the book, INLA works fine, but when I change a dataset, it keeps showing this error. The following is the code.
sdres <- sd(denim$waste)
pcprior <- list(prec = list(prior="pc.prec", param = c(3*sdres,0.01)))
formula2 <- waste ~ f(supplier, model="iid",, hyper = pcprior)
result <- inla(formula2, family="gaussian", data=denim)
result <- inla.hyperpar(result)
summary(result)
Error in str2lang(s) : parsing result not of length one, but 0
But another example works great.
sdres <- sd(pulp$bright)
pcprior <- list(prec = list(prior="pc.prec", param = c(3*sdres,0.01)))
formula <- bright ~ f(operator, model="iid", hyper = pcprior)
result <- inla(formula, family="gaussian", data=pulp)
result <- inla.hyperpar(result)
summary(result)
with result
"inla(formula = formula, family = \"gaussian\", data = denim)"
Time used:
Pre = 1.82, Running = 3.46, Post = 0.0998, Total = 5.38
Fixed effects:
mean sd 0.025quant 0.5quant 0.975quant mode kld
(Intercept) 6.977 1.015 4.98 6.977 8.972 6.977 0
Random effects:
Name Model
supplier IID model
Model hyperparameters:
Expected number of effective parameters(stdev): 1.00(0.002)
Number of equivalent replicates : 94.96
Marginal log-Likelihood: -366.67
Any hints would be appreciated!