Contact FutureLearn for Support
Skip main navigation
We use cookies to give you a better experience, if that’s ok you can close this message and carry on browsing. For more info read our cookies policy.
We use cookies to give you a better experience. Carry on browsing if you're happy with this, or read our cookies policy for more information.

Wrap up

This week we have learned different mathematical model types, their impact on the obtainable solutions and the limitations imposed by data availability. Compared to the Problem and Model steps of Week 3 these challenges are not economic per se but more methodological constraints we have to live with. Nevertheless, they feed back into our model design and thereby into the underlying economic question we want to answer.

Always aim to keep your initial numerical models small with little test cases that you could ideally reproduce with pen and paper. Most models can be gradually extended (ie first only introduce production constraints then add transport constraints and so on). This helps to ensure that the building block elements you add are working the way you expect them to, before going to the next element. Thereby, you may also figure out at what point your model becomes infeasible or starts to take much longer to solve. Designing a large numerical model and then trying to understand why it does not solve is much harder than going stepwise from small too big.

With more experience in modeling you will start to automatically include mathematical, computational and data restrictions into your initial problem assessment. This can be both helpful and hindering. On the one hand it often helps you to design a workable model for your problem right from the start. On the other hand it sometimes limits you too much in your ideas how to address your original research question. Be aware of this effect.

When it comes to handling all the results of numerical models you will need to develop skills in ‘number crunching’ and visualization. Basic statistical knowledge is helpful to generate meaningful aggregates of your obtained simulation results. But the most important part is that you know and understand the topic you analyse. Relating your results to the discussion within your research topic is key in ensuring that the reader gets the message. And it helps you figuring out which numbers are the important ones that need to be presented.

One final remark on numerical modeling: you cannot learn it just by reading, you have to do it. There is a lot learning by doing involved, especially with respect to your modeling software.

Share this article:

This article is from the free online course:

Exploring Possible Futures: Modeling in Environmental and Energy Economics

University of Basel