RJags Non-informative hyper-prioris Jeffreys prior?


I have a model where a random variable is normal distributed. And I wanna compute bayesian analysis with hyper-priors for the mean and the variance.

This is my model, but I know that it can't work because of the hyperparameters.
My question is how to compute this in rjags, if the variance is inf and the paramters of the gamma d. are both zero. I read about the Jeffreys priori für mu is constant 1 and for the inverse gamma for sigmaq is 1/sigmaq.

model_normal_combi1<- 'model {
for (i in 1:N) {
x ~ dnorm(mu, sigmaq)
mu ~ dnorm(0, Inf)
g ~ dgamma(0, 0)


If somebody knows a solution or a source for reading it would be great.

Thank you and have a good day


Not a robit
I have only used rjags once, so I am not to much help there. But I have a couple of questions, which may help others get involved and help me understand to better understand.

So you are running a linear regression model?
You have a predictor, which you want the prior to be 0, infinity?
By hyper-parameters you are referencing the mean and stdev or variance?
What does sigmaq stand for?