I am trying to forecast data using Bayesian analysis combined with a known general equation in R. I have set my priors and likelihood accordingly, and I am using the function ‘R2OpenBUGS’ to call an app (OpenBUGS) to run the Bayesian analysis. OpenBUGS usually carries out an analysis based on Bayesian MCMC + Gibbs Sampler algorithm. However, while plotting the quantiles of my final result, I presumably get the same plot for my 10% 50% and 90% quantiles.
This idea has been used before and it seems to work and so I am assuming that I am doing something wrong in my coding (maybe the prior distributions? maybe likelihood? or maybe the way I am extracting the results?).
If someone can take a look at my code, and give me some pointers. I will greatly appreciate it.
here is the code and the dataset: