« Why automation won't replace data scientists | Main | R 3.0.3 is now available »

March 06, 2014

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

With an opening slide like the one on the linked-to RUGS presentation, I can see why it's difficult to get more women interested in STEM fields.

-- one can gain some insight into the dynamics of the observable time series.

This is, generally, the notion that causes quants to crash economies. The common logical leap is to predict the future based on "figuring out" past data. While I'm not inclined to go all out Taleb, quant analysis of data generated by (mostly) humans making financial policy and decisions is dangerous. While one might be able to predict gross moolah movements, The Great Recession proved that most quants couldn't even figure that out. There were reasons that all that money moved to US housing, and all of them were the result of explicit policy decisions. The money flows happened as a result of the policies, not the other way round. Policy drove (and predicts) the data.

Financial markets are overwhelmingly driven by changes to the rules of the game, which changes are often hidden and nearly always carried out by those who've got much skin in the game. While The Great Recession, although not necessarily its exact date of occurrence (pick a day), was easily seen in advance by looking at the trajectory of house prices and their unsustainability, it was not obvious in less granular data. The old fashioned non-quant macroeconomists were the first to figure it out. Not that anyone would listen. "We all have to keep dancing while the music still plays."

IOW, identifying the existence of prior black swans is of little use to predicting future ones, since the manipulations driving past ones are, more or less, made illegal in their aftermath. Future black swans will be caused by nefarious manipulation in other parts of the financial landscape, and likely by other operators.

The financial system doesn't follow neat and clean algorithms a la Newton, thermodynamics, or even Heisenberg.

The comments to this entry are closed.


R for the Enterprise

Got comments or suggestions for the blog editor?
Email David Smith.
Follow revodavid on Twitter Follow David on Twitter: @revodavid

Search Revolutions Blog