# AdaptiveMCMC defining prior and multivariate parameters

**URL:** <https://discourse.julialang.org/t/adaptivemcmc-defining-prior-and-multivariate-parameters/93917>\
**Category:** Probabilistic Programming\
**Created:** [February 2, 2023, 11:46am UTC](https://discourse.julialang.org/t/adaptivemcmc-defining-prior-and-multivariate-parameters/93917 "2023-02-02T11:46:11Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![astro-kevin](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/astro-kevin/32/43342_2.png) [@astro-kevin](https://discourse.julialang.org/u/astro-kevin)\
**Post date:** [February 2, 2023, 11:46am UTC](https://discourse.julialang.org/t/adaptivemcmc-defining-prior-and-multivariate-parameters/93917/1 "2023-02-02T11:46:11Z")

</div>

I’m trying to make a start on making a MCMC chain using AdaptiveMCMC. I’ve figured out a function which returns a log-likelihood given a specific set of parameters (a vector of parameters). I’m not quite sure how this goes into the `adaptive_rwm` function because the function seems to require a `log_p` function which calculates the log-likelihood to have one set of inputs (which I presume to be the parameters). Do I just make my data into global variables and use them that way?

Also, how does one define your priors for the parameters? I have made priors using Distributions.jl, but I’m not sure how that applies to either the `adaptive_rwm` function or the `log_p` function.