Search found 3806 matches

by tomfid
Fri Jan 27, 2017 6:06 pm
Forum: Ventity
Topic: Mortality rates in a cohort model depending on creation time of the cohort
Replies: 28
Views: 60739

Re: Mortality rates in a cohort model depending on creation time of the cohort

I made a few mods, partly cleanup, but mainly to implement the switchover from cohort-specific mortality tables to a group generic mortality table. CohortPopGroups3 Test Swiss Population V2.zip That creates discontinuities in the mortality rate for each cohort, which seems problematic. It might be b...
by tomfid
Fri Jan 27, 2017 3:28 pm
Forum: Vensim
Topic: Lognormal in Vensim
Replies: 25
Views: 31804

Re: Lognormal in Vensim

Not at all dumb/boring. I can think of two ways to do a Monte Carlo with a lognormal distribution. Assuming your equation is: z = EXP( mean_ln_x + RANDOM NORMAL( -6,6,0,stdev_ln_x,seed )) 1. You can make the seed parameter a VECTOR input in the sensitivity setup. Or, use a seed of 0 and vary the NOI...
by tomfid
Thu Jan 26, 2017 6:43 pm
Forum: Vensim
Topic: Lognormal in Vensim
Replies: 25
Views: 31804

Re: Lognormal in Vensim

If the actual distribution of the data is lognormal, it seems odd to work with the ordinary mean and variance, rather than the geometric mean and variance. Be that as it may ... geometric mean = ln( arithmetic mean / sqrt( 1 + arithmetic variance / arithmetic mean^2 ) ) geometric std dev = sqrt( ln(...
by tomfid
Mon Jan 23, 2017 6:07 pm
Forum: Vensim
Topic: Tutorials (Video + text)
Replies: 19
Views: 43430

Re: Tutorials (Video + text)

There's a new series on discrete and stochastic events at
http://vensim.com/discrete-stochastic/
by tomfid
Mon Jan 23, 2017 6:02 pm
Forum: Vensim
Topic: Lognormal in Vensim
Replies: 25
Views: 31804

Re: Lognormal in Vensim

Oops! I goofed.

You want

z = EXP( mean_ln_x + RANDOM NORMAL( -6,6,0,stdev_ln_x,seed ))
or
z = geometric mean * EXP( RANDOM NORMAL( -6,6,0,stdev_ln_x,seed ))
by tomfid
Mon Jan 23, 2017 5:59 pm
Forum: Vensim
Topic: Somebody help me please!! ramp and pulse train combination
Replies: 33
Views: 14487

Re: Somebody help me please!! ramp and pulse train combination

The MODULO function is simpler - you might try it.

You can pass the sawtooth from MODULO through a lookup if you want to get a complex shape for the pulse train.

I think if we knew what you were doing, we could make better suggestions.
by tomfid
Mon Jan 23, 2017 5:46 pm
Forum: Vensim
Topic: Lognormal in Vensim
Replies: 25
Views: 31804

Re: Lognormal in Vensim

Suppose you have data x= 10 100 1000 Compute y = ln(x) = 2.3 4.6 6.9 Mean(ln(x)) = 4.6 and stdev(ln(x))=1.88. Note that exp(4.6)=100, as you'd hope. Then you can generate a sample via: z = mean_ln_x * EXP( RANDOM NORMAL( -6,6,0,stdev_ln_x,seed )) You can do all this in Excel using the average() and ...
by tomfid
Fri Jan 20, 2017 7:40 pm
Forum: Vensim
Topic: difficulty to download models from the models library
Replies: 9
Views: 5279

Re: difficulty to download models from the models library

It worked, but it was in my spam queue. Not sure why.
by tomfid
Fri Jan 20, 2017 7:07 pm
Forum: Vensim
Topic: difficulty to download models from the models library
Replies: 9
Views: 5279

Re: difficulty to download models from the models library

There are no comments in my queue on metasd.com. Possibly another browser problem?

I've realized that I overused the Poisson distribution in the car model, so a new version will be arriving soon.
by tomfid
Fri Jan 20, 2017 4:14 pm
Forum: Vensim
Topic: Somebody help me please!! ramp and pulse train combination
Replies: 33
Views: 14487

Re: Somebody help me please!! ramp and pulse train combination

1 - Every feedback loop must have a stock in it. This is not only a PLE limitation, it's also reality. A simultaneous equation loop might be a workaround for some "stiff model" situation, like a rapid market-clearing process embedded in a model with slow dynamics, but more often it's a con...
by tomfid
Fri Jan 20, 2017 3:30 pm
Forum: Vensim
Topic: difficulty to download models from the models library
Replies: 9
Views: 5279

Re: difficulty to download models from the models library

Works fine for me too. I think the MIME type for mdl files is set incorrectly, but if you right-click-download them, they work fine (otherwise they open as text in the browser).

The private message system on the SD society forum doesn't seem to be working.
by tomfid
Thu Jan 19, 2017 6:34 pm
Forum: Ventity
Topic: Mortality rates in a cohort model depending on creation time of the cohort
Replies: 28
Views: 60739

Re: Mortality rates in a cohort model depending on creation time of the cohort

That sounds sensible. The reason I'm interested in an explicit stock of health status is the following thought experiment. Suppose you have two initially identical cohorts. Cohort A starts around WWII and experiences the diet, exercise and health care that prevailed over the actual intervening decad...
by tomfid
Wed Jan 18, 2017 11:43 pm
Forum: Ventity
Topic: Minor question for CohortPop2
Replies: 1
Views: 2816

Re: Minor question for CohortPop2

I don't have it running at the moment, but that message indicates that an equation uses inputs that aren't shown on the diagram. This used to happen a lot in actions because there were some loose ends in the diagramming code. You should be able to edit Parent.Head Cohort, right-click the missing cau...
by tomfid
Wed Jan 18, 2017 11:41 pm
Forum: Ventity
Topic: Mortality rates in a cohort model depending on creation time of the cohort
Replies: 28
Views: 60739

Re: Mortality rates in a cohort model depending on creation time of the cohort

This is not my area, but I'd think it could get messy. Some thoughts: Frequently, models use mortality rate = f(age) implemented with a lookup. You could generalize to: mortality rate = f(age, gender, behavior, time) where the time component picks up the technical progress aspect. Substituting and e...
by tomfid
Wed Jan 18, 2017 11:23 pm
Forum: Vensim
Topic: Lognormal in Vensim
Replies: 25
Views: 31804

Re: Lognormal in Vensim

The wikipedia article on the lognormal distribution, https://en.wikipedia.org/wiki/Log-normal_distribution , lists formulas for converting between the arithmetic and geometric means and variances. If you have a sample of data, the natural thing to do is to compute the mean(log(data)) etc. directly. ...
by tomfid
Mon Jan 16, 2017 2:57 am
Forum: Vensim
Topic: Lognormal in Vensim
Replies: 25
Views: 31804

Re: Lognormal in Vensim

If you want x distributed lognormal, with mean(x) = 165, that's not what you have here. Your equation gives mean(log(x)) = 165. The mean of the lognormal is exp(mu + sigma^2/2), where mu and sigma are parameters of the corresponding normal. The median and geometric mean of the lognormal are exp(mu)....
by tomfid
Mon Jan 16, 2017 2:31 am
Forum: Vensim
Topic: Lognormal in Vensim
Replies: 25
Views: 31804

Re: Lognormal in Vensim

Your formulation suggests that some quantity has a geometric mean of e^165 - what physical meaning does that have? It's more than the number of particles in the universe.
by tomfid
Sun Jan 15, 2017 5:53 pm
Forum: Vensim
Topic: Lognormal in Vensim
Replies: 25
Views: 31804

Re: Lognormal in Vensim

In single precision Vensim (i.e. any version but DSS DP), taking EXP(x) with x greater than about 30 will cause an overflow. In double precision you can get to something like x ~= 300. But either way, anything that big is unlikely to be meaningful, so I'd guess you have an error. If you really have ...
by tomfid
Sun Jan 15, 2017 2:04 am
Forum: Vensim
Topic: How to simulate a Raleigh distribution with a Weibull one
Replies: 4
Views: 2839

Re: How to simulate a Raleigh distribution with a Weibull one

I agree that the stretch parameter is the one to use.

For all the functions, the result is essentially
y = shift + stretch*x
where x is some kind of standard version of the distribution.
by tomfid
Tue Jan 10, 2017 3:03 pm
Forum: Vensim
Topic: how to set a target value for my model?
Replies: 5
Views: 2708

Re: how to set a target value for my model?

Assuming the latter, you could do the following: - Create a metric that is a function of the gap between your target and actual values, like error = (target-actual)^2 weight = {some negative value} - Create a payoff that minimizes the metric *p error/weight - Optimize For this to work, the model and...
by tomfid
Tue Jan 10, 2017 2:22 pm
Forum: Vensim
Topic: how to set a target value for my model?
Replies: 5
Views: 2708

Re: how to set a target value for my model?

Are you looking for a dynamic target that affects the behavior of decisions, or goal-seeking optimization?
by tomfid
Mon Jan 09, 2017 3:53 pm
Forum: Vensim
Topic: Obtain current state of intermediate steps in Dynamic functions
Replies: 2
Views: 2255

Re: Obtain current state of intermediate steps in Dynamic functions

Some dynamic variables are implemented as macros, and you can turn on visibility of their internals (Tools>Options>Settings). Check "Show" for "Macro Variables" (near the middle of the dialog pane). However, the discrete or infinite order delays are not among them. If you need th...
by tomfid
Thu Jan 05, 2017 3:52 am
Forum: Vensim
Topic: Bug Report: Options Window in Model Reader v6.4E is Blank
Replies: 3
Views: 2228

Re: Bug Report: Options Window in Model Reader v6.4E is Blank

We won't revert. This is probably something simple. Thanks for the report.
by tomfid
Fri Dec 30, 2016 3:47 pm
Forum: Vensim
Topic: cumulative value over specified time range
Replies: 2
Views: 1903

Re: cumulative value over specified time range

Something like:

Code: Select all

cumulative = INTEG( inflow, 0 ) ~ $
inflow = STEP(1,start time) * fund_flow ~ $/day
start time = 5 ~ days