Skip to content
All library documents

Neural Kalman Filtering and Bollinger Bands for Pairs Trading

Article arXiv papers · Author: Amit Milstein et al.
Read the original

Summary

The document presents KBPT, a pairs trading policy that combines a Kalman filter, Bollinger Band signals, and a neural network based on the KalmanNet architecture. Traditional versions model the relationship between two assets as a linear state-space process with Gaussian noise. Because this model may only approximate the pair’s behavior, the resulting indicators and trades can be unreliable. KBPT instead uses an extended state-space model that allows for partial cointegration and learns to improve the tracking process.

Training occurs in two stages: first the tracking algorithm is tuned without a trading objective, then it is adapted to track indicators in a way that seeks to maximize revenue while approximating Bollinger Band decisions with a differentiable mapping. The authors report improved revenue over model-based and data-driven benchmarks across different assets. The description provides no asset list, evaluation period, transaction cost assumptions, or quantitative results, limiting assessment of robustness and live-trading relevance.

Key ideas

  • The method augments Kalman filter and Bollinger Band pairs trading with a KalmanNet-based neural network.
  • Its extended state-space model represents the asset relationship as partially cointegrated.
  • Training first tunes tracking independently, then adapts it to indicators and revenue objectives.
  • A differentiable mapping approximates Bollinger Band trading decisions.
  • The authors report higher revenue than model-based and data-driven benchmarks across several assets.

Tags

Pairs tradingMachine learningMean reversionTechnical indicatorsStatistics
Full text
# Neural Augmented Kalman Filtering with Bollinger Bands for Pairs Trading


# Neural Augmented Kalman Filtering with Bollinger Bands for Pairs Trading









Pairs trading is a family of trading techniques that determine their policies based on monitoring the relationships between pairs of assets. A common pairs trading approach relies on describing the pair-wise relationship as a linear Space State (SS) model with Gaussian noise. This representation facilitates extracting financial indicators with low complexity and latency using a Kalman Filter (KF), that are then processed using classic policies such as Bollinger Bands (BB). However, such SS models are inherently approximated and mismatched, often degrading the revenue. In this work, we propose KalmenNet-aided Bollinger bands Pairs Trading (KBPT), a deep learning aided policy that augments the operation of KF-aided BB trading. KBPT is designed by formulating an extended SS model for pairs trading that approximates their relationship as holding partial co-integration. This SS model is utilized by a trading policy that augments KF-BB trading with a dedicated neural network based on the KalmanNet architecture. The resulting KBPT is trained in a two-stage manner which first tunes the tracking algorithm in an unsupervised manner independently of the trading task, followed by its adaptation to track the financial indicators to maximize revenue while approximating BB with a differentiable mapping. KBPT thus leverages data to overcome the approximated nature of the SS model, converting the KF-BB policy into a trainable model. We empirically demonstrate that our proposed KBPT systematically yields improved revenue compared with model-based and data-driven benchmarks over various different assets.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.