On 12 July 2014, David H. Bailey and Jonathan M. Borwein (two of the bloggers on this site) presented the talk Scientific Integrity in Mathematical Finance at the Workshop on Optimization, Nonlinear Analysis, Randomness and Risk, held at the Centre for Computer-Assisted Research Mathematics and its Applications (CARMA), University of Newcastle, Australia. The viewgraphs for the talk are available here.
In this talk, Bailey and Borwein summarize research outlined in their paper (co-authored with Marcos Lopez de Prado and Qiji Jim Zhu), Pseudo-mathematics and financial charlatanism: The effects of backtest overfitting on out-of-sample performance. The talk also includes a series of 24 graphics illustrating attempts to achieve an optimal back-tested investment strategy, based only on a pseudorandom time series. These graphics were generated by a computer program due to Stephanie Ger and Marcos Lopez de Prado.
Similar material was included in a talk Big data computing: Science or pseudoscience?, presented to the Australian Mathematical Sciences Institute National Seminar Series on 11 July 2014. This talk discussed issues of mathematical finance in the greater context of reproducibility and scientific rigor in “big data” computing.