…Inside The Black Man's World


November 19, 2012

Implicit Model vs Explicit Models


This phrase stuck in my head when I was at the World Money Show. It was right after Ernest Chan finished up his speech on “The Pitfalls of Backtesting”. You can find his blog here and his books here. (Just like to add Ernest Chan is a really nice guy!)
While I was sitting, a couple passed me and I overheard them say “forget about the pitfalls of back testing I have never even heard of back testing.”

So I looked at the World Money Show timetable again and noticed that they were correct. That no presenter decided to discuss the world of back testing. Granted that there was a chat about the pitfalls of back testing but nothing about back testing itself. I have talked about the benefits of back testing in a post called “Automated Trading Strategies.” So I’m going to speculate on the reasons that the couple never heard of back testing before;

1) Listening to Gurus’.

2) They are using an implicit model for trading or investing.

So let’s go with number one, Gurus! At the World Money Show after a lot of the presenters were finished speaking a number of people just wanted to ask them where this item was going or where that item was going. I have said this many times! I don’t believe anyone can predict the future. This is no discredit to analysts, traders or investors who use predictive models. If someone is giving their analysis on what is going on “right now” that is different.

An Implicit model is “where the assumptions are hidden, their internal consistency is untested, their logical consequences are unknown, and their relation to data is unknown.[1]” Traders or investors who trade on instinct and their gut feelings use implicit models. As a opposed to Explicit models, “where assumptions are laid out in detail, so we can study exactly what they entail”. [2]

The fact is that you can take an Implicit model and write it down on paper. A lot of traders have some sort of model they use to form their opinion and investment decisions. When comparing the benefits of I.M. to E.M. you can see which one has more benefits.

The problem is, as soon as you have an Explicit model, people assume that you can predict the future. People assume that once the explicit model has been established that the model has predictive capabilities.

On FRI, 2008 APR 11 | 07:45 AM ET on CNBC David Harding of Winton Capital explained his explicit model, but as soon as it was established that he had an explicit model the question of predictability came into play.

Joe Kernen : “Hence when can you quickly give me your best picks, long and short, that your computers are indicating right now in some of the major markets?”

David Harding: “Well, you know, i can’t give you best picks. We don’t have reliable ways of forecasting markets. We succeed by just going on and we have only a slight edge.”

It seems that Explicit models are always thought of as having predictive abilities. This all goes back to back testing which the couple mentioned, as you can see from the chart above there are too many positives in Explicit models, as long as they aren’t used for predictive purposes.

[1]Epstein, Joshua M. (2008). ‘Why Model?’. Journal of Artificial Societies and Social Simulation 11(4)12 http://jasss.soc.surrey.ac.uk/11/4/12.html

[2]Epstein, Joshua M. (2008). ‘Why Model?’. Journal of Artificial Societies and Social Simulation 11(4)12 http://jasss.soc.surrey.ac.uk/11/4/12.html

About the Author

Trend Movements is the brainchild of Tahric Finn and Chad Grant. Both men are active traders in the markets, and are willing to discuss financial market trends and beliefs at any time. They believe we should all have some knowledge of how finances work. Even you! Ya you! The guy (or gal) reading this right now!




Be the first to comment!

Leave a Reply