Ventana software support forum

No. I'm not sure what exp(events) would mean. In general, a likelihood (or log likelihood) is f(model,data). For the standard normal, f = (model-data)^2/2, which is why the payoff computes a sum of squares. The idea is the same for Poisson, just the functional form is different. I'd suggest copying ...

Right. It's in the model. Because there isn't a Poisson calibration payoff element in v6.4, you have to calculate a variable that represents the Poisson log likelihood. Then you use that as a policy payoff, and maximize it. I wouldn't expect the difference between Normal and Poisson to be large - it...

It's linked to the 3rd post on the third page of this thread.

Binomial makes sense if the number of errors a driver can make is limited to a finite set each month. That seems unlikely. I think you want Poisson. Both distributions will be built into v7 (any day now), but otherwise, see the original model I posted, illustrating the Poisson distribution.

You can call it whatever you want of course. The software doesn't care. However, conceptually, the reason is just what you say - "Is it not the rate when the other variables are held equal to zero?" The initial error rate will be equal to exp(B0) only if all the other vars are initially ze...

I think you want to think of exp(B0) as the _reference_event rate, not the _initial_event rate. Then make it a parameter to be optimized, not derived from the first month of driving.

I can't remember how the export treats NA values, but it's possible that you could create an extra variable:

x no 0s = IF THEN ELSE( x = 0, NA, x )

and export that. The NAs might get skipped or left blank.

Speaking of dummy ranges, one way to handle the proliferation of dimensions would be to create a 2D x 2D function. Then users could insert a dummy dimension ( one : one1 ) on one side when they wanted 2D x 1D, etc. Making the number of loops a parameter might also be possible, but tricky in the comp...

1) set the minimum to a small value, like 0.001 < x. However, if your time step is 1 unit, the shortest sensible time constant for anything is 1 unit (or >2 units if you want a decent approximation of continuous time).

But, back to your real question, it's definitely possible. It just requires a lot of planning, because every time we introduce something like this, it has side effects on a lot of infrastructure, like the external function system.

1) Just run the model with those parameters to see what the floating point error is. Then fix the structure so it can't happen. https://www.vensim.com/documentation/23320.htm 2) Seems like a bad idea. One month of driving will be a noisy measure. Better to use the mean, which you can get with INITIA...

This is actually not a bug, because it's the specified behavior of the function, but I agree that it would be useful. I think what we actually need is a more general linear programing approach to such things. That can be done with the lp_alloc link, http://vensim.com/workbench/ , though with signifi...

3) The model you attached is the same as my original concept, so I don't have any comment on that. Just as a matter of good source control practice, I'd recommend keeping your whole project in a folder with a version number, and periodically copying the whole thing to a new folder with an incremente...

What do you think?

I think you are correct.

Recall my example above: stock = INTEG( flows, initial stock ) initial stock = 3.141 In the same way, rewrite your stock equation so that the 200 is a separate variable, "initial miles driven," with value 200 (or whatever). Then use that constant in your optimization setup. To remove <time...

You have to add the constant to the model first - edit your equation so that the stock initial conditions are named constants. Then you can use those constants in your control file. You can delete a variable with the Delete tool in the toolbar, or in the Equation Editor, using the Delete Variable bu...

If you want the initial conditions of the stocks to be included, make them parameters: stock = INTEG( flows, initial stock ) initial stock = 3.141 Then add the "initial ..." constant to the optimization control file. Using a weight of 1 is probably a good start, as long as driver errors ar...

It appears that you're trying to load a dataset that doesn't exist. Check what you have specified in the Data Sources field of the Advanced tab of the Simulation Control dialog.

I think the problem is not the stock structure, but the accounting. Judging by Profit/Loss, the mine is already losing money in year 2, so it's no surprise that it shuts down. A couple thoughts: - You might want to calculate an expectation of profitability that's separate from cash flow. The trouble...

At least your simple example should be easily accomplished with DELAY PROFILE. https://www.vensim.com/documentation/index.html?fn_delay_profile.htm You could put a table of any distribution into the profile lookup, though this would still be somewhat inconvenient for cases where the parameters chang...

The loop here is pretty simple: cash flow affects Produksi -> Revenue -> etc., and ultimately determines itself. This is conceptually impossible. There has to be a stock in the loop somewhere. I'm not sure what's happening here, but my guess is that Produksi actually depends on a stock of accumulate...

The problem is just that "Invoker" is a reserved word. Change it to "caller" or "invoculator" or whatever, and it runs.

There will definitely be more.

You're right. It's not a bug. I reverted to 6.4E and it works fine. I get: Writing 6 values for -events per month[Driver60] Writing 6 values for -events per month[Driver57] Writing 6 values for -events per month[Driver56] Writing 6 values for -events per month[Driver54] Writing 6 values for -events ...

... or maybe more than 2, if you want to keep track of males and females separately. I don't think that' necessary for a first cut though.