CoMSES Net Discourse Forums

Should I trust that model? (2016)

First questions: trust the model to do what? Models can be claimed reliable for different purposes (Edmonds et al. 2019). Prediction is one of the most difficult and if you want this then you should only trust a model when it has had a track record of correctly predicting unknown stuff (NOT the same as just fitting known data). See what Nate Silver and his team do when really trying to predict (Silver 2012).

Edmonds, B., le Page, C., Bithell, M., Chattoe-Brown, E., Grimm, V., Meyer, R., Montañola-Sales, C., Ormerod, P., Root H. & Squazzoni. F. (2019) Different Modelling Purposes . Journal of Artificial Societies and Social Simulation, 22(3):6. http://jasss.soc.surrey.ac.uk/22/3/6.html. (Open Access)

Silver, N. (2012) The Signal and the Noise: Why So Many Predictions Fail-but Some Don’t. Penguin.