Abstract
We introduce an original application of Suprathreshold Stochastic Resonance (SSR). Given a noise-corrupted signal, we induce SSR in effort to filter the effect of the corrupting noise. This will yield a clearer version of the signal we desire to detect. We propose a financial application that can potentially reveal big positions in a market. We assume there exist return signals that correspond to big orders, which are hidden by noise from small scale traders. We induce SSR in an attempt to reveal these big positions.
Highlights
Stochastic Resonance (SR) is a phenomenon in which a weak periodic signal in a non-linear system is amplified and achieves a noise induced maximum [1]
Our goal is to reveal this autocorrelated stream of small orders that correspond to large orders
Recall our assumption that there exist large orders that are broken into small orders with autocorrelated execution times
Summary
Stochastic Resonance (SR) is a phenomenon in which a weak periodic signal in a non-linear system is amplified and achieves a noise induced maximum [1]. The addition of white noise to a sub-threshold periodic signal can increase the ability to reveal this signal. Benzi suggested that Earth’s climate falls in one of two stable states, either glacial or interglacial. Nicolis showed that stochastic fluctuations in climate can trigger a state switch [5]. These state switches do not correspond with the periodicity of glacial periods. Benzi proposed that weak periodic astronomical forcing caused by variations in Earth’s orbit, coupled with short term stochastic climate variations resonate, and the coupling of these events result in periodic state switches. Benzi’s generalization of this effect is that a dynamical system that is subject to both periodic forcing and random perturbation may achieve a peak in power spectrum when the two resonate [6]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
