Statistical Analysis Stack Exchange
Statistical Analysis Stack Exchange

Questions tagged [rstan]

Relating to the R bindings for mc-stan

41 questions
15
votes
4 answers

How best to deal with a left-censored predictor (because of detection limits) in a linear model?

Context: I'm new to Bayesian stats and am trying to fit a multiple regression with rstan. All variables are continuous and there is no hierarchical structure. One of my predictors is left-censored because it falls below the detection limit for a…
mkt
  • 11,770
  • 9
  • 51
  • 125
6
votes
1 answer

Why do model selection (AIC and LOO) outcomes differ between ML and bayesian approaches

I am interested in understanding whether my continuous data (dput code at bottom for reproducibility) are fit better by a linear model (Gaussian distribution) or a gamma distributed model. I typically use the lme4 package in R (maximum…
Dylan_Gomes
  • 177
  • 12
5
votes
2 answers

Are Jacobian adjustments necessary when the target parameter is a difference between two parameters in Stan?

[Note on cross-posting: This question has now been posted on the Stan Forums as well.] I want to model the index called Delta P (e.g., p.144 of this paper), which is basically a difference between two proportions (i.e., $\frac{n_1}{N_1}$ -…
Akira Murakami
  • 263
  • 2
  • 9
4
votes
1 answer

Why is a Gelman-Rubin diagnostic of < 1.1 considered acceptable?

In multiple sources a Gelman-Rubin MCMC convergence diagnostic of less than 1.1 is considered evidence that chains have converged. For example in this…
user3322865
  • 41
  • 2
3
votes
1 answer

Nonlinear sin model with brms

I try to fit sin function with brms using next code: library(tidyverse) library(brms) N <- 100 x <- seq(0, 10, length.out = N) e <- rnorm(N) y <- 7 * sin(2.5 * x) + e inp_data <- tibble(x, y) plot(inp_data, type = 'l') priors <- prior(normal(7,…
user2579566
  • 174
  • 1
  • 7
3
votes
1 answer

hurdle model with non-zero gaussian distribution in R

I have biomass data (continuous response variable). If sufficient data is collected, the log(Biomass) follows a normal distribution. However, I am separating the overall biomass by family (i.e., biomass for each family) and in some sites no families…
Jess
  • 33
  • 1
  • 3
3
votes
0 answers

Is there a way to use the brms function bayes_factor without having to refit models to not have prior samples?

In the brms package the function bayes_factor allows you to compare two models. Personally, I prefer the WAIC or LOO methods of comparing, but it is helpful to have both these and bayes factors for those who like bayes factors for these purposes. I…
BKV
  • 380
  • 2
  • 10
3
votes
1 answer

What exactly does it mean that a 90% Credible Interval is computationally stable?

I was reading the rstanarm documentation and came across this about its use of 90% intervals as the default. I was hoping someone might be able to provide some clarification. Default 90% intervals We default to reporting 90% intervals rather than…
BKV
  • 380
  • 2
  • 10
2
votes
0 answers

How to define informative priors from previous studies using stan_glm?

I am trying to develop a linear regression model for estimating stature from handprint measurements. I would like to employ the Bayesian approach and define informative priors from the previous studies. I have a data set with several predictors…
ivanJ
  • 21
  • 1
2
votes
2 answers

Cross validation on a single model (not model comparison)

I understand the method of cross validation to be to leave out some part of a dataset (whether that be one data point at a time = LOO, or subsets = K fold), and train the model on some data, test the model's predictive accuracy with the remaining…
Dylan_Gomes
  • 177
  • 12
2
votes
2 answers

Need help understanding hurdle model specification and results interpretation

I am trying to use hurdle gamma model for one of my use cases, to handle a zero-inflated scenario. I have a very simple code creating dummy data with quite a few zeros. # Dataset prep non_zero <- rbinom(1000, 1, 0.1) g_vals <- rgamma(n = 1000,…
Srivatsa Srinath
  • 61
  • 6
2
votes
1 answer

Modelling Parameter $r = \max\limits_{i = 1, \dots , 10} p_i - \min\limits_{i = 1, \dots , 10} p_i$ of Binomial Random Variable in Stan/RStan/R

I'm trying to use Stan and R to fit a model that, uhh, models the observed realisations $y_i = 16, 9, 10, 13, 19, 20, 18, 17, 35, 55$, which are from a binomial distributed random variable, say, $Y_i$, with parameters $m_i$ (the number of trials)…
The Pointer
  • 1,064
  • 13
  • 35
2
votes
0 answers

stan - 2 approaches to missing value imputation; which is better and why?

So, me and a colleague have to impute some data, x, given a categorical variable. We arrived at two different approaches: a) as in the tutorial: split x into x_obs and x_mis, and treat x_mis as parameters. Something like this: data { int…
user2089357
  • 448
  • 2
  • 9
1
vote
1 answer

How to write proportionality in a Stan model?

I am having difficulties in writing the following equation into a Stan model. $$ y_i = \mu(x_i) + \epsilon_i \\ \epsilon_i \sim N(\theta,\sigma^2) \\ \mu(x_i) = a + b x_i \\ p(a) = p(b) \propto 1 \\ \sigma^2 \sim Inv-Gamma(0.001,…
John
  • 11
  • 1
1
vote
1 answer

Help with rstan models

I would need help in order to write a specific Stan model. The biological question The idea of the model is modeling the number of Bones (NbBones : discret continuous count) according to Orders (3 different discret orders) and Time (continuous).…
chippycentra
  • 193
  • 6
1
2 3