Jim Thorson

It would be great to add an identiability check for Q_ik (and possibily X_xtp). This would involve inferring the intercepts from the `RhoConfig` and then column-concatenating the two, taking the...

#### Description: `obj$report()` doesn't seem to work when `MakeADFun(..., intern=TRUE)` #### Reproducible Steps: It appears that when `intern=TRUE`, the `obj$env$random` and `obj$env$last.par` are no longer visible. This then appears to...

Importance sampler for mixture models?

2

I'm hoping to implement a state-space model that includes Markov-switches between "regimes", which represent different parameter values for the function representing the expected value of changes over time. This model...

Delta-method for expectation of derived variables

2

Hi all, It appears that sdreport() uses the delta-method to approximate the variance of derived values (specified as ADREPORT() outputs), but does not use the delta-method to approximate their expected...

difference between make_install.R and devtools::install_github?

3

Hi all, I confused myself recently by getting in the habit of installing github via the devtools function: devtools::install_github("kaskr/adcomp/TMB", ref="v1.3.0") where ref="v1.3.0" refers to a stable release. This advice is...

Measure of normality for random effects from importance sampling

1

Hi all, I just re-tried using the importance sampler (option MCcontrol in MakeADFun, via internal function MC()), and see that it now reports the standard error of sampling -- very...

Working with eigendecomposition tools in TMB

6

Hi all, I've been looking through the functions distributed with TMB, but can't seem to work out how to use the options for computing eigendecomposition. Is there any example code...

Anyone working on AD variational Bayes?

1

Hi all, Given Cole Monnahan's success replicating Hamiltonian MCMC into TMB (based on the NUTS algorithm from STAN)), I was wondering if anyone was interested in continuing to explore STAN...

Generalized hypergeometric function?

1

Hi all (especially @skaug), There's many sampling designs that are reasonably analyzed using an "N-mixture" model, and recent work (http://onlinelibrary.wiley.com/doi/10.1111/biom.12521/full) shows a closed-form solution for this likelihood involving a generalized...

Automatic detection of numerical overflow?

1

Hi all, I have a model where I'm reasonably sure that I've diagnosed the following problem. Namely, this model (and others I've seen) given a certain configuration would end up...