Most Recent Stories

1,000 Years Worth of Market Data

I love this post by Tadas at Abnormal Returns asking the following:

“If we had a 1,000 years of market data what kinds of things would get validated? What things would lack support in the data?”

There are so many good responses in there. But here’s what I said:

“1,000 years worth of market data would certify that the financial markets did something in the past and that no one knows precisely what that means for the future.

Imagine that – with 1,000 years worth of market data we’d be able to perform some pretty reliable tests. We’d probably be able to confirm much of what we already know. But it would also open the door for a lot of misinterpretation of the data and people who sell the past as a reliable indicator of the future. In my opinion, the future is going to be less predictable than we think due to two elements:

  • Our financial lives are necessarily short because they tend to be a series of short-terms inside of a long-term. And the sh0rt-term is really, really difficult to predict.
  • The financial system is an adaptive system that is always changing. The system of the future will always be something completely different from the system of the past.

This is why I stress how skeptical we have to be using past performance data and backtesting. No matter how much historical data we have the future is still going to be uncertain. So, at the end of the day, the best we can do is use the historical evidence we have and interlace it with a sound operational understanding of our world to produce what are likely to be high probability results.