Ventana software support forum

Re #1, you can press CMD + G to find the other instances of the search. I think it might be missing from the menu as ‘Find Next’, and it would be nice to have it explicitly in the menu bar under 'Edit' (although we have 'Find Again', but the term and shortcut might be confusing). In addition, it wou...

I just saw this without Kalman. So, it might not be even related to the number of datasets (though it happens every 16 back to back runs). It is not critical, and I will let you know specifically what the reason for it is if I find how to reproduce it.

When running optimization+Kalman, does Vensim turn off the filter and run the model with the estimated parameters obtained from Kalman to see if there is still a good fit even without state resetting?

Thanks, Tom. The workaround I had worked perfectly, but CLEARRUNS should speed it up even further. I've never seen this behavior when running MCMC without filtering (on single and MC) so I thought it could be related to Kalman (on single core). Thanks again.

I'm running a script like this: SPECIAL>NOINTERACTION SPECIAL>LOADMODEL|MG_Kalman.mdl SIMULATE>DATA|data_v01.vdfx SIMULATE>RUNNAME|Powell_kalman_v01 SIMULATE>OPTPARM|Powell_Settings_kalman.voc SIMULATE>PAYOFF|Payoff_kalman.vpd MENU>RUN_OPTIMIZE|o MENU>EXIT SIMULATE>DATA|data_v02.vdfx SIMULATE>RUNNAM...

Since there's no information about this in the documentation, what does 'Too many iterations. Optimization not complete' mean?

I'm running Kalman on Market Growth and I get FP error right before the optimization starts. The error is caused by state resetting, but it's not related to the discussion we had earlier, https://www.ventanasystems.co.uk/forum/viewtopic.php?t=8553, which already helped a lot. The overflow happens be...

This makes a lot of sense. Thank you so much, Tom!

Thanks a lot, Tom! Very helpful! I have a few other questions: 1. Should I also use 'PN1Stock | S/ COVAR' too? Or, is 'S/1e-10/1e-10' enough? 2. In general, when should we actually use 'Lev_1 | Lev_2/ LEV_1_2_COVAR'? When there are two process noise inputs? 3. How would you define 'LEV_1_2_COVAR'? B...

When there's a mismatch between parameter names in the model and the ones in the tab file, Vensim only throws the 'Problems encountered in the simulation.' message. Could you please include a more informative message like 'Parameter name mismatch between model and tab'? Thank you. Screenshot 2025-07...

This is similar to the model I've shared earlier (https://www.ventanasystems.co.uk/forum/viewtopic.php?t=8552). I've noticed that state resetting is applied to all stocks, not just the one I define in the '.prm' file (PN1Stock/sp1_var/init_sp1_var). In these situations, how do I prevent a random sto...

Please run a Powell with X/RParallel vs RRandom on the attached model. The parallel search picks values beyond the parameter boundaries, which is really confusing. I don't have any explanation or workarounds for it, and I'd appreciate any inputs on this. Thank you! Kalman X R Parallel.zip Screenshot...

I have a clarification question. What is the exact payoff contribution factor calculated with Overdispersed Poisson? I'm trying to understand the underlying function to avoid making incorrect assumptions about the standard deviation, etc. Thanks a lot!

Here's the code in case anyone needs it. You can modify it to open in the background whenever Vensim is open or you should manually open it whenever you're working with workflows like mine. Again, it needs to be compiled to get an executable binary (took me only a few minutes to implement the whole ...

ChatGPT helped me write a helper to use Accessibility “AXObserver” API to receive window-created notifications so that whenever this window pops up the OS automatically minimizes it. it's a swift app that you can compile with terminal command line tool and run it in the background, which uses less t...

They're so sticky that none of these work

As someone who goes back and forth between different inputs and datasets when working on a model, I appreciate the fact that this message pops up every time I run the model. In this way, I can make sure that I've run what I intended to do. But the window management is done so poorly, and it has been...

Thanks so much for adding this option! One can easily define a shortcut through macOS System Settings. I would've assigned ⇧⌘R but it seems to be reserved (hard coded) and it acts like ⌘R. 'Show in Window Explorer' is also available on Windows. Very cool! Screenshot 2025-07-05 at 8.42.43 PM copy.png...

I've been running some experiments using the CMD script (https://www.ventanasystems.co.uk/forum/viewtopic.php?t=8538) and the model (https://www.ventanasystems.co.uk/forum/viewtopic.php?t=8526) I've shared recently. I noticed that on Vensim multi-core, MCMC gets stuck entirely randomly on different ...

Thank you, Tom. Unless I'm missing something, I haven't figured out how to generate .out files through dll yet:
viewtopic.php?t=8521

I just noticed that even if I'm doing it manually, whenever I close Vensim window I need to recompile the model. so maybe it's the expected behavior. Edit: I realized that I can write everything in a single CMD (without having 'No_Interaction' or 'Load_Model') to avoid recompiling the model. Somethi...

I have this script, but when I run it back to back with different {idx} it compiles the model from scratch every time, which is confusing because i'm not changing anything in the model (like changing the constants in between the iterations). If I repeat the steps manually inside Vensim by changing t...

I have some basic questions about Kalman setup. I've generated a synthetic data, using the same model, and I'm trying to use this payoff: *CK model|data/variance Which contributes to: - (model-data)^2/variance/2 - LN(variance)/2 I think I should use EVM1 as the variance even though I have only 1 ser...

I understand it now. Thanks a lot for the clarification!

This is a Lotka-Volterra model with an over-dispersed Poisson as payoff (I've shared a version of this model earlier). What does Vensim do when the event data is non-integer? I get these warnings at the very end of the estimation log: Thank you!