X-Y plot not X-t

[majidshafieejood]

I was just wondering if it is possible to plot one variable vs another in Vensim? For example all the time when I want to plot x-y graph, I need to export the data set and plot it in excel! In dynamic analysis, sometimes we need the plot of two variables to get see hysteresis!

thanks,

Majid,

[Administrator]

Using custom graphs you can set the x axis to something different.

http://www.vensim.com/documentation/ind ... custom.htm

[majidshafieejood]

Thanks

[majidshafieejood]

Hi again,

I have another problem hoping you can help me with that.

I have different plots. some of them are plots of state variables (say X or Y vs Time) versus time. Others are plots of state variables versus each other (say X vs Y). Is it possible to have different time spans for these plots. I mean for example for the first group of plots, I would like to have a plot starting from time=0, but for the second group, I want to ignore the first few steps (the plots starting from time=50).

Thanks,

[Administrator]

x-min and x-max in the custom dialog definitions will let you do this.
http://www.vensim.com/documentation/ind ... ?23930.htm

[tomfid]

Another possibility is to create reporting variables that suppress a portion of a variable's history. For example:

reported x = IF THEN ELSE( Time >= threshold, x, :NA: )

[majidshafieejood]

The First reply is not what I am looking for. But the second one is a very wise idea.
Awesome.
Thanks

[majidshafieejood]

Hello again,

I wanted to send a new post but I could not! So I am following my previous post.

I wanted to generated Gaussian white noise in Vensim (I am using PLE but I don't think it matters!)

So basically I need to use Random Normal distribution. Now the question is if in this function, the randomly generated values are uncorrelated or not. Since I can handle mean zero and finite standard deviation in Vensim but I couldn't find anywhere that the randomly generated values are uncorrelated. This is crucial for any type of white noise.

Thanks,

Majid,

[tomfid]

They are uncorrelated, as are draws from any of the RANDOM functions.

Completely uncorrelated white noise is actually not physically realizable, because it would have infinite power at high frequencies. So, you either need to ensure that your noise is sampled at some fixed interval, or implement a cutoff frequency, as in pink noise. See http://models.metasd.com/pink-noise/

One way to ensure fixed sampling is to not change your TIME STEP, but that prevents tests for numerical stability. Better to use SAMPLE IF TRUE with PULSE TRAIN or the PINK NOISE macro above.

[majidshafieejood]

Thank you very much.