back to list

Project: Efficient Granger causality

Description

Granger causality is among the standard functions for quantifying causal relationships between time series (e.g., closing prices of stocks). However, naïve computation of Granger causality requires pairwise comparisons between all time series, which comes with quadradic complexity. In this project you will focus on developing an algorithm that can find the pairs of time series with high causality more efficiently. The goal is to have a scalable and efficient system for identification of causalities over streaming data.

Prerequisites: ability to write efficient code in Java or Scala, successful completion of 2AMD15, comfortable with mathematical proofs.

Details

Supervisor

Odysseas Papapetrou