Models splinter when you look at them closely. While this may not be true of fashion models, Emmanuel Derman argues that it is true of financial and economic models, and over three posts I’ll explore what models are and some of their benefits and risks. As someone with a physics background, I feel a kinship with the physicist-turned-financial engineer as he writes and rants in Models. Behaving. Badly. His focus is finance and economics, but many of his general observations apply equally well to most types of models.
[My interest lies in the use of computer simulation models to improve performance in non-technical applications. (What exactly is non-technical I will leave vague.) As I pursue my studies and continue my business, I’m exploring and sharing a range of resources from the academic to the mainstream.]
Derman’s book is relatively mainstream, and I’ll start by saying that I don’t particularly recommend it unless you already have an interest in modeling. It’s a weird combination of anecdote, history, philosophy, and diatribe. Having said that I want to reflect on and expand on several of the topics he tackles.
What are models?!
Models serve a specific purpose
There are many types of models.
Physically, we build model airplanes that capture shapes and features. These models do not perform key functions of airplanes such as flying, enabling transport, etc. It could be argued whether a remote-controlled airplane is a model of a real airplane or whether it’s just a small real airplane. What do you think? Most toys and games are models of one sort or another.
We use fashion models to allow us to imagine what clothes will look like when worn (by someone unlike ourselves!)
A person posing as a model for a painter isn’t really a model if it’s the actual person who is being painted. The model in this case is the painting, the representation that doesn’t contain all the features of the real thing. (There are other possibilities that are tempting to pursue but I just don’t have time!)
In science, we rely every day on weather models. These models contain mathematical and computational representations of water, air, and heat flows and stores. Predictions are made by linking observations based on equations with roots in underlying validated theories of physics, chemistry, and biology.
Socially, we speak of role models, people who display or represent specific behaviours or patterns of behaviour that we view as desirable.
In business, various models represent the flows and stores of money, materials, people, and information are successfully used to design organizational structures, financial controls, business processes, and information systems.
Derman talks about economic models containing unobservable factors, immeasurable variables, and abstractions. This is an important point – we need to consider what proxies are available to observe the unobservable, measure the immeasurable, and test abstractions in the concrete world.
I’m going to leave you hanging to some degree and end the post here with the summary: there are a lot of different types of models that we use for different purposes, but in general we are emphasizing some features and eliminating others that are not important for a specific purpose.
See you next time in part 2.
Interested in models or have comments? Let me know at firstname.lastname@example.org.
And sign up for the newsletter!
Disclosure: I have no personal or business relationship with Derman or his book.
Reference: Derman, Emmanuel. Models. Behaving. Badly. Why Confusing Illusion with Reality Can Lead to Disaster, on Wall Street and in Life. New York. Simon and Schuster Free Press. 2012.