Page 1 of 1
DICE model
Posted: Wed Oct 15, 2008 4:13 pm
by duilio
Hi Bob, all
apologize for bothering,
I've got the DICE Model from Tom Fiddaman's web site but for some reasons I wasn't able to run it. It seems that it cannot "see" the external data, but the data are there.
Can you help me?
Thanks in advance!!
G
Posted: Thu Oct 16, 2008 7:10 am
by Administrator
In Vensim, click "Model->Import Dataset".
Select the file "NORDDAT2.DAT" and click "Open".
You might get a message "Dataset NORDDAT2 already exists. Do you want to overwrite it?". Click "Yes".
Hopefully you will get the message "Conversion completed without error".
Now click on "Model->Simulate".
Click the "Advanced" tab.
Click the "Data Sources..." button and select "NORDDAT2.VDF".
Click the "Simulate" button.
Hope this helps. The model and all files were prepared a long time ago so file formats have probably changed since then.
Tony.
DICE Model
Posted: Thu Oct 16, 2008 8:21 am
by duilio
Hi Tony,
that's exactly what I did.
But I've got two error messages:
ERROR: The variable Emiss Stabilization - and perhaps others require a data file be specified
ERROR: Unable to correctly load in exogenous values - aborting
Should I have to re.build the model from scratch?
In this case is there anybody who can give some hints (at request, when I'll have questions...)
Thanks in advance
G
Posted: Thu Oct 16, 2008 9:25 am
by Administrator
Has the data imported correctly?
The error suggests you have missed the following steps.
Now click on "Model->Simulate".
Click the "Advanced" tab.
Click the "Data Sources..." button and select "NORDDAT2.VDF".
Tony.
DICE Model
Posted: Thu Oct 16, 2008 10:00 am
by duilio
Hi Tony,
I did it, I've added the data source on the advanced tab, the error is still there.
It seems there is something corrupted in my version, maybe problem of Windows, it can be everything...
Btw I'm trying to replicate the model from scratch, we will see what happens.
Thanks
G
Posted: Thu Oct 16, 2008 10:11 am
by Administrator
Did the dataset import correctly?
I followed those steps again just to check if I had missed anything, it worked fine for me.
Tony.
DICE MODEL Again: I need help
Posted: Mon Nov 10, 2008 10:18 am
by duilio
Hi Tony, all
I had to re-build the model from scratch, I did all the necessary steps you described but the model did not work. Maybe I lack some fixed to the software. Anyway, Now I have a working model. I still have to add the equations that use subscripts.
So I need help in understanding the use of subscript sin this model. What 's the meaning of the equations in the subscript (T1-T10) ? Particularly what is the purpose of the T subscript in this specific model?
Apologize for asking a narrow question, but I need to understand it...
Thanks a lot in advance
G;)
Posted: Mon Nov 10, 2008 10:43 am
by bob@vensim.com
The T subscript is used to implement policies at different times (ie 1965, 1985, 1995, 2000...). The model uses SHIFT IF TRUE to move through the different times - eg
Shift Red = SHIFT IF TRUE(GHG Red Fracs[T1],shift switch=1,T10,0,GHG Red Fracs [T1])
This is a clever, but confusing, way to implement this. It could more easily be done with VECTOR LOOKUP but that function was not available when the model was written.
DICE Model
Posted: Mon Nov 10, 2008 1:28 pm
by duilio
Thanks!!
G
DICE Model, last question for Today...
Posted: Mon Nov 10, 2008 3:33 pm
by duilio
Guys,
If I want to optimize the Discounted Utility which type of contraint should I have to set - up in the payoff definition?
Is it a contraint strictly needed for this optimization?
Thanks a lot!
G
Posted: Tue Nov 11, 2008 10:34 am
by bob@vensim.com
Constraints do not go in the payoff function but in the range for the search variables. For example if you are searching over a faction it will usually need to be between 0 and 1. For the payoff you just need discounted utility with a weight of 1. The payoff computation accumulates this.