Posted by & filed under Behavioral Finance.

The last financial crisis has hit the market landscape like never before and we believe that financial calibration of models has its share of responsibility in the fueling of Boom & Bust cycles.

First things first, by “no financial calibration” we mean not changing the parameters of a model to better fit the economic context it faces.

What are the benefits? Well, we believe this is the only way to always have a stable benchmark to refer to.

Imagine what would happen if, every now and then, someone at the “Bureau International des Poids et Mesures” in Paris decides to change the length of the etalon meter: the entire metric system would crumble, and people would lose the perception of distances. One day it takes you 500 meters to walk to the baker, the next day that same distance becomes 700 meters, two years later 300, and so on.

Financial Calibration: metre etalon

Straightforward right? Well, it didn’t seem so when authorities implemented Basel II in the banking system to restrain credit overexposure. The accord actually ended up encouraging Banks to take more and more risks in the name of “know your client” motto.

Why? Because between 2004 and 2007, Central Banks interventions to cure the DotCom crisis (which was a financial and not an economical one) fooled the models and led them to adapt to a “dream environment”. During this time, “old bankers” were sent to early retirement because they were too “conservative”, and banks continued pumping their balance sheet and feeding the next crisis.

On the other side, if you stick to a stable and not calibrated model, you can rather focus on improving the interpretation of the data (according to the previous example, you can walk 500 meters, or half a kilometer, or 0.31 miles), while keeping your feet on the ground because your basic parameters stay exactly the same!

This is how the Quantesys algorithm, by avoiding financial calibration, was able to predict the 2008 crisis, as well as the March 2009 rebound.

On the other side, while it perfectly predicted the April 2010 sell-off, it failed to predict the September rebound. The year was “flat”, but there was a silver lining: we knew exactly how and why it happened, and what the reaction was going to be.

Something external had tainted the model (ie. Central Bank interventions most likely), preventing the market to go further down. Fine, this is exactly what we mean by improving the interpretation of existing data rather than changing the model. This one time failure gave a clear signal that the market would have been heading to its historical heights, with the intention of breaking them eventually.

What would have happened by applying financial calibration? Well, we would have maybe twisted the model to “explain” the September 2010 anomaly, but not understood why, three years later, S&P500 would be breaking new highs.

Thanks to this prediction, Quantesys client could ride the wave all the way up to new highs.

Fnancial Calibration: quote

Alberto Ravandoni

Marketing & Communication at Quantesys
To effectively capture the attention, you have to create valuable content, spread it toward the right audience at the right moment and with a pinch of irony.