investment model, need help plz

[almoon]

Hi All,

I am working on a university model,

The purpose of this model is to study the best choice of investment for a university manager.
the university should invest in facilities or increase number of available staff,

these decisions should help in increase probability of offer rejection rate.

There is some problem which am trying to fix, but in general can you give me your feedback about the work in general and help me to fix the problem ?

Thank you very much.

[Administrator]

There is some problem which am trying to fix, but in general can you give me your feedback about the work in general and help me to fix the problem ?

What is the problem?

I'd really spend the time and make sure you add units to each of the parameters in your model. For example, INVESTMENT SHARE probably needs to be dimensionless.

Personally, I don't believe any results from models that have unit errors. They are the first check that any good model will pass.

[LAUJJL]

Hi Almoon.

Your model is too complicate to start with and it is full of unit errors.
Start with something simple that you can work with and show to some people.
Joined a simple model that works only on staff. It works with 'strictest unit testing'.
You can choose something else, or build different simple models like this one based on different point of views, for instance based on facilities only.
Then you can step by step make your model more complex at the condition that you have understood the previous ones.
The model may be bugged, I built it in half an hour, but it does not really matter, it shows the way.

Regards.

JJ

[almoon]

Thank you very much for your help, but what about the student population part?

[LAUJJL]

Make a model with only students, that manages only the student side: how long take the studies, what makes them choose an univeisity, the fees, the job they can expect to find etc...
You can after having got a good understanding of simplified models, start to mix different factors, but with care and with plenty of time with no rush.
Work too with people who know about university management too. Only those can guide you towards making something useful to them.
Regards.
JJ

[Administrator]

Thank you very much for your help, but what about the student population part?

What is your question on the student population?

[almoon]

Thanks admin, My question whether is part of the model is working ? is it good to start with ?

[Administrator]

Only you can decide if the model is doing what you want it to.

I would personally say no, as you need to put units into your model so that it passes basic checks.

[Denis]

How it will be improve ?

[tomfid]

It's always possible to build a model without units, and adding units won't in itself improve the model.

However, in my experience grading hundreds of papers at MIT, it's extremely unlikely that a model will be good the first time. The process of adding units will reveal many conceptual inconsistencies that can then be corrected. It doesn't guarantee a good model, but it helps.

Units also help others to interpret the model. I don't like to look at models without units, because it's so hard to sort out conceptually what is going on.

Tom

[LAUJJL]

Hi Tom.

What you consider as evident and actually is (puting units in models) does not seem evident for many people and especially for the papers published in the SD conference with model joined in Vensim, where about 50% have either no units, or have unit errors and are published by academic people! I personnally do not even read such papers that are difficult to understand and difficult to believe in, the authors not knowing the basic concepts of SD. In the next conference, I do not remember either having seen a model with mass balance checking, something that is basic too.
What do you think about these facts?
See you at the conference.

Regards.

JJ

[tomfid]

It's a somewhat sorry state of affairs. I have to read a lot of unitless papers, because I'm a thread chair. I'd like to summarily reject them in most cases, but that would lead to an unfair asymmetry, because it's easier to judge the quality of a model provided in supplemental material rather than merely described in text.

[tomfid]

The key point for me is that it's easier to build a good model with units than without. Many people regard adding and checking units as an extra step, neglecting the huge savings in effort from avoided errors and improved clarity that it can bring.

[LAUJJL]

Hi Tom

Concerning this problem, I think that papers should be published to be read.
And I do not read papers with a model without units or with unit errors, because I think that the author is may be competent in his domain but certainly not in SD.

When I started studying SD years ago, I too felt uneasy with units, but I very quickly (about a month) realized the interest of putting units.
People publishing models without units or with errors, are generally working with other tools (most of them) that do not do unit checking and follow their standard habit. It proves that they have very little experience in SD modeling, and I personally think that learning SD (both building model and analyzing them which is quite different) is a very long process. The deduction is that a model built without some basic principles will be difficult to understand and probably not interesting by itself or the way it is analyzed and I will not lose my time studying them.

I think then that one of the motivations of people publishing models is either not so much to be read, but adding material to their curriculum vitae, or they largely underestimate the difficulty of SD modeling.

One of the problems that faces the association is up to what point it will accept this situation.
Because if it rejects bad papers, the number of people attending the conference may drop considerably, something that SD people fear above all.
The present strategy is to my opinion good within a short horizon, but bad within a longer one.
It gives a very bad image of SD in general, especially to people who may be interested in the field, but will be bothered by such incoherencies.
You remarked that unit checking saves a huge amount of time. It proves that people that do not do that, just do not have any experience in SD. And the question is: is it reasonable to accept papers about SD, written by people with no experience with SD?

Such incoherencies do a lot of harm to the SD community, because non SD and SD people will always have a doubt about anybody pretending to have some knowledge in SD.

Best regards.
JJ