Kalman Filters for Equity Pricing and Financial Time Series
Summary
This discussion points readers to examples of Kalman filtering in finance, including an application to the Kyle model and a pricing framework for spot and futures markets. In the Kyle setting, a market maker updates an estimate of an asset’s final value as trades arrive, accounting for the possibility of informed trading. The cited spot-and-futures paper is described as a step-by-step Excel example, while other references cover R implementations and software packages.
The answers also suggest considering unscented Kalman filters and particle filters for financial time series that may depart from normality. The document is a collection of recommendations rather than a worked analysis: it provides no model specification, data, or performance results. The suitability of each filter depends on the problem’s assumptions, and the claim that alternative filters are more applicable is not supported with comparative evidence here.
Key ideas
- The Kyle model illustrates how a market maker can update an asset value estimate as successive trades arrive.
- Kalman filtering is presented as a tool for pricing applications involving spot and futures markets.
- The discussion points to Excel and R materials for learning and implementation.
- Unscented Kalman filters and particle filters are suggested for financial series with non-normal behavior.
- The document offers references rather than empirical results or a detailed implementation.
Tags
Full text
# Kalman Filter Equity Example # Kalman Filter Equity Example I am looking out for some material where I can study about Kalman Filter applied to Equity using Excel or R? ## Answer by Andrew (score 10, accepted) https://quant.stackexchange.com/a/4764 A great example of kalman filtering is in the Kyle Model. I have attached a presentation on the application of R to the kalman filter in the Kyle Model. http://www.rinfinance.com/RinFinance2009/presentations/microstructure-tutorial.pdf Basically in the Kyle Model, a market maker finds the likelihood an asset is ending up at a certain price given that a person is an informed trader. Given this, you update what the final price will be by each successive trade through a kalman filter ## Answer by Matt Wolf (score 14) https://quant.stackexchange.com/a/4703 A simple google search should get your started: I like this one the best because it compares different packages: - https://www.stat.berkeley.edu/~brill/Stat248/kalmanfiltering.pdf and here couple more: - http://www.r-bloggers.com/the-kalman-filter-for-financial-time-series/ - http://cran.r-project.org/web/packages/dlm/index.html - http://cran.r-project.org/web/packages/FKF/index.html - http://cran.r-project.org/web/packages/KFAS/index.html - http://cran.r-project.org/web/packages/schwartz97/index.html - http://www.jstatsoft.org/v41/i04 But I highly recommend you to also read up on unscented kalman filters and particle filters because they are much more applicable to financial time series (handle non-normality): - http://en.wikipedia.org/wiki/Kalman_filter - http://signal.hut.fi/kurssit/s884221/ukf.pdf - http://en.wikipedia.org/wiki/Particle_filter - http://user.uni-frankfurt.de/~muehlich/sci/TalkBucurestiMar2003.pdf - http://perso.uclouvain.be/michel.verleysen/papers/ffm07sd2.pdf - http://www2.mccombs.utexas.edu/faculty/carlos.carvalho/teaching/lopes-tsay-2010.pdf ## Answer by vonjd (score 6) https://quant.stackexchange.com/a/15715 The following paper gives you a step-by-step presentation of how to use the Kalman filter in an application in a pricing model framework for a spot and futures market. Everything is explained using Excel: A Simplified Approach to Understanding the Kalman Filter Technique by T. Arnold, M. Bertus and J. M. Godbey
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.