Does successful backtesting against old stock market data make your newest method safe?
Backtesting is the process of testing a trading strategy on relevant historical data to ensure its viability before the trader risks any actual capital. A trader can simulate the trading of a strategy over an appropriate period of time and analyze the results for the levels of profitability and risk. – Investopedia
Backtesting looks smart and it can prevent huge errors, but
Lawyers say that when the judge presents his judgement, everything before the but, is bovine scatology. So it is in this case.
The key is “relevant historical data.” With something as complex as the stock market there may be a good fit to old data, even a very good fit, but never a conclusive fit.
There will always be things you don’t know.
Hindsight is 20-20.
As always, the past is instructive about the future, but never conclusive.
Daniel Khaneman, one of the minds behind behavioural finance, says it like this:
“What hindsight does is it blinds us to the uncertainty with which we live. That is, we always exaggerate how much certainty there is. Because after the fact, everything is explained. Everything is obvious. And the presence of hindsight in a way mitigates against the careful design of decision making under conditions of uncertainty.”
In simple terms, you can never have a surprise using hindsight.
Foresight is not 20-20. Not even close.
No matter how good your data is and how good your plan is, it does not, even cannot, include everything possible. We make decisions with incomplete data.
The future will contain things that have not happened yet. The relative weighting of importance among the variables will be different. You will not be perfectly disciplined and organized.
Long Term Capital Management failed in 1998
Its staff owned upwards of 100 PhDs. Two of them held Nobel prizes in Economics. You can find an instructive article about it on Wikipedia. It was sudden and complete. No gradual disappearing for these folks.
Something happened that “could not” happen. A decisive end to be sure.
You will live in the future and it will be different
You cannot devise a perfect investment strategy based on the past because you cannot know everything that matters. Always keep that thought in mind. There is no perfect trading plan.
AI and robot based investing have the same constraints. People with natural intelligence program robots and they don’t know everything they should. By definition it is impossible to include surprises in any such program. Nonetheless, these programs may be an improvement over some of the amateurish choices available today.
Trading and investing are different
AI may be a perfectly good trading platform, but as an investing platform, the jury is still out. The long term approach to investing widens the scope for surprises.
Be cautious. Do not treat AI as a sure thing.
Don Shaughnessy arranges life insurance for people who understand the value of a life insured estate. He can be reached at The Protectors Group, a large insurance, employee benefits, and investment agency in Peterborough, Ontario. In previous careers, he has been a partner in a large international public accounting firm, CEO of a software start-up, a partner in an energy management system importer, and briefly in the restaurant business.
Please be in touch if I can help you. firstname.lastname@example.org 866-285-7772