# bayesian

**URL:** https://discourse.pumas.ai/c/estimation/bayesian/8.md

[Latest](https://discourse.pumas.ai/latest.md) · [Categories](https://discourse.pumas.ai/categories.md)

---

## [About the bayesian category](https://discourse.pumas.ai/t/about-the-bayesian-category/72)

<div class="topic-metadata">

**Author:** [@vijay](https://discourse.pumas.ai/u/vijay)\
**Replies:** 0

</div>

---

## [Updating the posteriors in the bayesian analysis](https://discourse.pumas.ai/t/updating-the-posteriors-in-the-bayesian-analysis/943)

<div class="topic-metadata">

**Author:** [@ahmed.salem](https://discourse.pumas.ai/u/ahmed.salem)\
**Replies:** 1\
**Last updated:** [September 22, 2023, 12:12pm UTC](https://discourse.pumas.ai/t/updating-the-posteriors-in-the-bayesian-analysis/943 "2023-09-22T12:12:46Z")

</div>

Hello, I developed a Bayesian PKPD model based using one data set. I collected another data set and I want to test the capability of the model to predict the new data set using two scenarios: 1- predicting the new data s…

---

## [Converting variance-covariance matrix to LKJCholesky prior in Bayesian estimation](https://discourse.pumas.ai/t/converting-variance-covariance-matrix-to-lkjcholesky-prior-in-bayesian-estimation/870)

<div class="topic-metadata">

**Author:** [@Sooyoung.Lee](https://discourse.pumas.ai/u/Sooyoung.Lee)\
**Replies:** 8\
**Last updated:** [May 23, 2023, 4:47am UTC](https://discourse.pumas.ai/t/converting-variance-covariance-matrix-to-lkjcholesky-prior-in-bayesian-estimation/870 "2023-05-23T04:47:24Z")

</div>

Hello, I am trying to perform Bayesian estimation. Then, I would like to use an informative prior distribution for the variance-covariance matrix based on LKJCholesky. However, I have been failed so far. I would like …

---

## [Error for BIC, AIC calculation from a Bayesian output](https://discourse.pumas.ai/t/error-for-bic-aic-calculation-from-a-bayesian-output/858)

<div class="topic-metadata">

**Author:** [@ahmed.salem](https://discourse.pumas.ai/u/ahmed.salem)\
**Replies:** 5\
**Last updated:** [April 11, 2023, 3:39am UTC](https://discourse.pumas.ai/t/error-for-bic-aic-calculation-from-a-bayesian-output/858 "2023-04-11T03:39:13Z")

</div>

Hello, I am trying to calculate BIC and AIC using the following code split\_method= LeaveK(K = 1) split\_by= ByObservation(allsubjects = true) res\_t = Pumas.truncate(res; burnin=100) pl = loglikelihood( res\_t ; split\_…

---

## [Error while running simobs using Bayesian fit](https://discourse.pumas.ai/t/error-while-running-simobs-using-bayesian-fit/841)

<div class="topic-metadata">

**Author:** [@ahmed.salem](https://discourse.pumas.ai/u/ahmed.salem)\
**Replies:** 8\
**Last updated:** [March 1, 2023, 8:24pm UTC](https://discourse.pumas.ai/t/error-while-running-simobs-using-bayesian-fit/841 "2023-03-01T20:24:45Z")

</div>

Hi I ran the following model BAYESIAN\_PKPD = @model begin @param begin tvKTV1 ~ LogNormal(log(0.4), 0.5) tvKTV2 ~ LogNormal(log(0.1), 0.5) tvecmo ~ Normal(-0.7, 0.5) tvCL1 ~ LogNormal…

---

## [\`truncated\` versus \`Constrained\` functions](https://discourse.pumas.ai/t/truncated-versus-constrained-functions/839)

<div class="topic-metadata">

**Author:** [@ahmed.salem](https://discourse.pumas.ai/u/ahmed.salem)\
**Replies:** 10\
**Last updated:** [March 1, 2023, 12:29am UTC](https://discourse.pumas.ai/t/truncated-versus-constrained-functions/839 "2023-03-01T00:29:36Z")

</div>

Hello, I am reading through the Bayesian workflow and I noticed that when the distribution is assumed to be within certain domain either Constrained() or truncated() functions are used. I am wondering if both functions …

---

## [Error with Bayesian Model Normal distribution for omegas](https://discourse.pumas.ai/t/error-with-bayesian-model-normal-distribution-for-omegas/840)

<div class="topic-metadata">

**Author:** [@ahmed.salem](https://discourse.pumas.ai/u/ahmed.salem)\
**Replies:** 6\
**Last updated:** [March 1, 2023, 12:28am UTC](https://discourse.pumas.ai/t/error-with-bayesian-model-normal-distribution-for-omegas/840 "2023-03-01T00:28:45Z")

</div>

Hello I have the following Bayesian Model BAYESIAN\_PKPD = @model begin @param begin tvKTV1 ~ LogNormal(log(0.4), 1.0) tvKTV2 ~ LogNormal(log(0.1), 1.0) tvecmo ~ Normal(-0.7, 1.0) tvC…

---

## [Bayesian model fails to estimate](https://discourse.pumas.ai/t/bayesian-model-fails-to-estimate/811)

<div class="topic-metadata">

**Author:** [@storopoli](https://discourse.pumas.ai/u/storopoli)\
**Replies:** 1\
**Last updated:** [January 10, 2023, 11:14am UTC](https://discourse.pumas.ai/t/bayesian-model-fails-to-estimate/811 "2023-01-10T11:14:42Z")

</div>

I have a Bayesian model that fails to converge. Here is the model. A 1-cmt model with proportional error and LogNormal priors and likelihood: @model begin @param begin tvcl ~ LogNormal(log(1.1), 0.25) …

---

## [Predict function after Bayesian estimation](https://discourse.pumas.ai/t/predict-function-after-bayesian-estimation/771)

<div class="topic-metadata">

**Author:** [@ahmed.salem](https://discourse.pumas.ai/u/ahmed.salem)\
**Replies:** 4\
**Last updated:** [October 5, 2022, 12:40pm UTC](https://discourse.pumas.ai/t/predict-function-after-bayesian-estimation/771 "2022-10-05T12:40:31Z")

</div>

Hello I am using the predict or simobs functions to sample from posterior distribution for each subject after using Bayesian estimation as following code: Random.seed!(12345) sims = vcat(\[DataFrame(predict(result\_PKPD;…

---

## [Bayesian Model Initial Parameters](https://discourse.pumas.ai/t/bayesian-model-initial-parameters/726)

<div class="topic-metadata">

**Author:** [@Rahulub3r](https://discourse.pumas.ai/u/Rahulub3r)\
**Replies:** 2\
**Last updated:** [July 20, 2022, 9:52am UTC](https://discourse.pumas.ai/t/bayesian-model-initial-parameters/726 "2022-07-20T09:52:17Z")

</div>

How does Pumas calculate initial parameter of the omegas in a Bayesian model for Inverse Gamma distribution ? For e.g. if I define an omega like this: ω²SCR0 ~ InverseGamma(10/2, 0.145\*(10 + 1 + 1)/2) and run init\_pa…

---

## [Post-processing steps for full Bayesian model](https://discourse.pumas.ai/t/post-processing-steps-for-full-bayesian-model/730)

<div class="topic-metadata">

**Author:** [@Rahulub3r](https://discourse.pumas.ai/u/Rahulub3r)\
**Replies:** 6\
**Last updated:** [July 19, 2022, 2:51am UTC](https://discourse.pumas.ai/t/post-processing-steps-for-full-bayesian-model/730 "2022-07-19T02:51:05Z")

</div>

I am using the following code from the Pumas tutorials website for fitting a Bayesian model for theophylline Codeusing Pumas, MCMCChains theopmodel\_bayes = @model begin @param begin # Mode at \[2.0, 0.2, 0.8, …

---

## [Individual PK Parameters following Bayesian Estimation](https://discourse.pumas.ai/t/individual-pk-parameters-following-bayesian-estimation/665)

<div class="topic-metadata">

**Author:** [@ahmed.salem](https://discourse.pumas.ai/u/ahmed.salem)\
**Replies:** 11\
**Last updated:** [May 23, 2022, 11:34am UTC](https://discourse.pumas.ai/t/individual-pk-parameters-following-bayesian-estimation/665 "2022-05-23T11:34:53Z")

</div>

Hello, I am trying to run inspect function following Bayesian estimation to get the individual random effects but I am getting the following error ERROR: type BayesMCMCResults has no field approx Stacktrace: \[1\] getpr…

---

## [Incorporating Inter-Occassion Variability in Bayesian Model](https://discourse.pumas.ai/t/incorporating-inter-occassion-variability-in-bayesian-model/674)

<div class="topic-metadata">

**Author:** [@ahmed.salem](https://discourse.pumas.ai/u/ahmed.salem)\
**Replies:** 3\
**Last updated:** [April 25, 2022, 6:18pm UTC](https://discourse.pumas.ai/t/incorporating-inter-occassion-variability-in-bayesian-model/674 "2022-04-25T18:18:07Z")

</div>

Hi, I am trying to incorporate between-occasion variability in Bayesian Model as following model\_bayes\_IOV = @model begin @param begin θ ~ Constrained( MvNormal( \[2.68, 5.47\], \[1.0 0.0; 0.0 …

---

## [Unable to specify a correct prior distribution](https://discourse.pumas.ai/t/unable-to-specify-a-correct-prior-distribution/371)

<div class="topic-metadata">

**Author:** [@anqipan](https://discourse.pumas.ai/u/anqipan)\
**Replies:** 15\
**Last updated:** [November 19, 2021, 3:12pm UTC](https://discourse.pumas.ai/t/unable-to-specify-a-correct-prior-distribution/371 "2021-11-19T15:12:52Z")

</div>

Hi, I am having difficulty specifying a prior distribution in Pumas with lower bound and initials, my codes is: @param begin θ ~ MvNormal(\[16.6,13.1,4.05,4.89,0.623,0,0,0\], Matrix{Float64}(\[0.0895 0.…