Ventana software support forum

Hi, Is there any way to do integration over a variable that is not related to TIME STEP, i.e. time-invariant? For example, P(mu) is a probability density function of mu where mu is the time-invariant variable which has a predefined range, say, 1 to 10. Now I want to integrate variable A over the dis...

Hi Tom, No, it is the model without the priors. Only MLE, no Bayesian. I have reflected on this issue a bit more. Perhaps I was fundamentally mistaken by expecting the mean of the error to be 0. This case is not linear, and hence in general the mean of error would not necessarily be close to zero? I...

Hi, I want to ask a question about good practice in handling missing data points in Excel data imported into Vensim, in the context of calibration. Suppose the following is what I want to calibrate: Time 1 2 3 4 5 6 7 8 9 10 Excel data 5 6 -- 5.5 7.3 -- -- 4.5 7.8 9.5 Model 5.3 5.7 5.2 6.0 6.9 3.8 6...

Hi Tom, thanks for your reply. Yes, it is weighted error. The weight is the standard deviation - calculated from the optimization and then plugged back into the model to replace the original weight of 1. The resultant log likelihood is very close to half of the number of data points, namely, around ...

Hi, Regarding optimization through MLE (fitting model to data), what if, after getting the true log likelihood by using the standard deviation of error term as the proper weight, it is found that the mean value of the error term is not very close to 0? Is it a violation of the fundamental assumption...

Hi Tom, thanks, however I cannot understand what you meant by "value changed on the last pass through the parameters". Please could you elaborate a bit further? Many thanks.

Hi Tom, Sometimes the optimization result comes with a star (*) ahead of a parameter being optimized. Something like the following: Maximum payoff found at: *Inventory of products = 5e+007 What does the star (*) mean? I remember there are some old posts in which people asked the same question, howev...

Hi Tom, I have been trying to understand why the payoff likelihood follows a Chi squared distribution with 1 degree of freedom . I found several posts including this one and some related to calibration, where you offered very useful insights and clarifications. I just wanted to ask some further ques...

Hi Tom, I have given more thoughts to your advice on using Binomial or Poisson for likelihood function, rather than Gaussian for likelihood function. According to Payoff Element Types in Reference Manual, for Binomial, the payoff contribution is: log( model p success^data_n_successes * (1-model p su...

Hi Tom, thank you so much for your advice, all of which is very enlightening. I did read one of you articles which was regarding dynamic cohorts and I found it very inspiring. Actually I knew my model about the scrappage is very much over-simplified but I just wanted to use it to serve as a context ...

Hi Tom, thanks for your advice. I have tried to do it in 2 steps. The 1st is a regular optimization using Powell through which I found calibrated parameters at maximum payoff. Then, the 2st starts from this optimum status (by replacing the original parameters with the calibrated parameters in the mo...

Hi, If I want to use MCMC as a method of payoff sensitivity analysis in the standard Powell optimizer, do I have to manually edit the .voc file to set the options associated with MCMC? I am asking because when using the user interface (after clicking Optimizer and going to Optimization Control panel...

Many thanks for sharing the links Tom. They are useful. Meanwhile, I would be very grateful if you could take a look at the model and offer some advice (please see my questions above). Thank you so much for your time.

Hi Tom, thank you so much for your further reply. As mentioned yesterday, I have attached a slightly revised model which should reflect the (hypothetical) application context better than the model yesterday. Basically the situation is that I have a dataset representing vehicle inventory as a functio...

Thanks a lot for your quick response Tom. Highly appreciated. The iteration to estimate the SD is for the errors, not for the prior. I did not make myself clear enough. But, the point is, after getting the SD for the errors, it will need to be used for the Total log likelihood which is the sum of th...

Hi Tom, many thanks for this model and your explanation. It well illustrates the relative role of priors. My question is: what is the consideration behind the formula for Data Weibull, i.e. true scale * ( -LN( RANDOM UNIFORM(0,1,0) ) )^(1/true shape) ?

tomfid wrote: ↑Mon Sep 17, 2018 7:03 pm Try the attached: Weibull calibration 2.mdl

Hi Tom, thank you so much for your detailed reply. My further questions are below, highlighted in blue colour: A couple things: In the attached model, I still use the explicit expression of the log likelihood (instead of the mixed payoff that you mentioned), just to help myself better understand the...

Hi Tom, Thank you so much for your reply. In a hypothetical case, such as the rate of identifying some flaws in products, I want to calibrate my model against a set of measurement data whose distribution follows Weibull distribution ("Data Weibull PDF"). So I define my model as a Weibull d...

Hi Tom, I know this is an old post, but I find it very useful to me as someone just beginning to learn how to apply Bayesian approach to calibration of parameters in Vensim. Having carefully read through the above discussion and in particular your replies, I still have some question about creating a...

Thanks a lot Administrator. Please could you explain why a Sum function is needed here?

Thanks.

Administrator wrote:You cannot suppress the error (as it is an error).

One alternative would be to do something like I've done in the attached model.

Hi, When using VECTOR ELM MAP function, would it be possible to avoid the error message (due to mapping outside the range) and set those :NA: to be some value such as 0? Basically I want to "shift" the values of a subscripted constant A to a subscripted constant B, by 5 subscript units. So...

Thank you both so much, Tom and LAUJJL !

Hi Tom, Many thanks. I did not know there is a NORMAL CDF function available now in Vensim. Just found in my Vensim (version 7.1). But it does not appear to be covered in Vensim Help? Please could you let me know where more details about this function. Thanks. Note that the NORMAL CDF macro is now b...

Hi Administrator, Many thanks for your quick response. Yes I did come across the second one, i.e. https://www.ventanasystems.co.uk/forum/viewtopic.php?t=3438 . So, it appears that using a closed-form expression to approximate the CDF is the way out? I am concerned about the accuracy of the results o...

Hi Tom and everyone, This might be trivial. However I searched the forum and did not find the answer to it. I am trying to model the cumulative distribution function (CDF) and hazard function of certain distributions that do not have analytical expressions (closed form solutions). The background is ...