Dr. Mikhail Kopytine

My current research is focused on probabilistic forecasting of financial time series, elucidating and quantifying the informational redundancy of the dynamic "context" which is the financial markets. I am creating software for using such redundancies for wealth generation via algorithmic trading. The editorial policy I've created for a dynamic web site promoting such services emphasizes white-paper research reports, and I am a regular author of such reports.

Joined an international collaboration recreating conditions of the early Universe at a US national lab. Using wavelet transform, revealed modification of correlation structure of particle production in nuclear collisions. Led group efforts in a hypothesis-driven search for interpretation of these results, using analysis and simulations. In cooperation with the leading expert, worked out an algebraic connection between own wavelet-based measure of correlation and an autocorrelation function.

Co-taught graduate physics lab. Analysed particle production in high energy nuclear collisions in a CERN collaboration experiment. Initiated a novel scalable technique of correlation analysis, based on discrete wavelet transform. Convinced community of its merit, which required an OO design and a C++ implementation of the algorithms to synthesize and analyze correlated patterns, along with a detailed analysis of correlated noises related to the instrumentation.

nuclear physics, experiment, data analysis, statistics, wavelets