This post is inspired by an article in Wired Jan-14 - Why Quants Don't Know Everything
Don't quants know everything?
Whilst in finance quants are blamed for too much dependence of numerical patterns, the discipline of quantitative analysis is entering unexpected sectors: sports, journalism, culture, … and even politics.
Does the rise of the quants happen in four stages? Pre-disruption - disruption - overshoot - synthesis?
Let me jump into the overshoot stage:
The most profound example of overshoot, of course, happened in the finance industry, where the rise of quantification could concentrate decision making - and moneymaking - within a relatively small group of people at a bank's headquarter.
And once they are empowered, quants tend to create systems that favor something pretty close to cheating.I have touched this issue in Stop Lying With Models, replying in When Three Rights Make a Wrong.
NN Taleb - "The Black Swan" - has famously shot down the notion of "risk" to an excuse of quants to sleep well.
The main question: how can you get insight into something (that is not predictable) without scenario analysis based on quant methods? What if you have no idea what the optimal risk may be?
Synthesis is the stage of co-existence, co-creation and interaction between computers and humans for better decision making.
But I interprete the 4 stages in the article as: time before bid data take over (pre-disruption) - big data analytics takes over (disruption) - too much dependence on data (overshoot) - the synthesis of big data and human intuition.
But I believe The Big Joke of Big Data is not only relevant to finance, but also other fields. And ironically NN Taleb delivers the most important argument (from his deep knowledge in statistics - an I add based on my experiences in machine learning).
And what is not touched in such discussions at all: models are not only operational - they speak about qualitative dependencies and behavior (in the sense of the philosophy of speculative reality they are reality.
So, IMO, quants don't know everything, but much.
Picture from sehfelder