CPD Events
Detecting And Predicting Price Jumps With Mahalanobis Distance And Signatures
11 Sep 2024
About the event
This study extends Jump Models for dynamical systems by sampling observations from a distribution to estimate hidden states. It links the detection of financial time series price jumps with anomaly detection in time series segments. A new three-step method is proposed: 1) jump detection using Mahalanobis distance with path signatures; 2) training Jump Models with this indicator; and 3) predicting hidden states from new data. The approach enhances accuracy by identifying outliers post-jump; with analysis on simulated data showing the method's effectiveness in accurately retrieving true hidden states without relying on future information.
Send an enquiry
Fitch Learning
Part of the Fitch Group, Fitch Learning partners with clients to enhance knowledge, skills and conduct. Fitch Learning is a global leader in training with experience of delivering specialised technical training at all levels to the financial community. Fitch Learning partner with clients to elevate knowledge and skills and enhance conduct. We work with 9 out of 10 of each of the largest Investment Banks, Asset Managers and Global Banks and through state-of-the-art training centres in London, New York, Hong Kong, Singapore and Dubai, and our leading distance learning portals, we train more than 20,000 delegates each year.