The Princeton – UChicago Annual 7 Nov 2014 Quant Conference
Mukul Pal speaking on Data Universality
Why should an investment analyst need a special skill set different from that of a sports analyst, a beverage company executive, a scientist, a social change agent, or the key decision makers in the government. Data, and its interpretation, is what connects us all despite our domains. We may not always be able to reconcile sub-atomic data with stock market data as both of them vibrate at a different frequency, but that does not change some universal laws which are present in every data set, irrespective of its natural source. Statistical laws are not only at the heart of physics, but also drive how we look at data. This means that there is the overlooked influence of ‘Data Universality’. Universality can be defined as “The aspects of a system’s behavior which are independent of the behavior of its components. And even systems whose elements differ widely may nevertheless have common emergent features”, Therefore, Data Universality can be defined as the “common universal behavior of any data set irrespective of its organic source of generation or derivation.” The talk on Data Universality will explain the disconnect between Pareto’s 80-20 and Galton’s Mean Reversion; how investing styles can be simplified; how Capital Asset Pricing Model can be transformed into a framework for modeling growth and decay of any natural variable driven by data.