Generational Regimes
Buttonwood from the Economist recently published an interesting article describing the essence of generational regimes, summarizing results from the Barclay Capital 2010 Equity Gilt Study (27 Feb print edition, p. 82). Although seemingly irrelevant for short-term trading, understanding generational regimes is crucial for two reasons:
- Regime self-similarity: generational regimes illustrate that finance is broadly self-similar across all investment horizons, from intra-day to inter-generational
- Context: generational regimes are to trading what cosmic background radiation is to astrophysics, providing long-term fabric on which all short-term oscillations overlay
Generational regimes are another root cause why Naïve Backtesting is Bogus, as assuming stationarity and ergodicity across generational regimes is economically nonsensical. Yet, despite this fact, methodologies for adjusting quantitative backtesting for generational regimes remain an interesting open research question.
Memetic Portfolio Optimization
While researching multi-objective optimization for large search spaces and clustering, Quantivity stumbled upon some interesting recent work by Aranha and Iba applying memetic and biologic algorithms. Latter work appears along the line of Brabazon and O’Neill Biologically Inspired Algorithms for Financial Modelling (which, responding to previous comments, may be potentially useful for those looking for published examples of GA applied to trading, as it includes numerous simplistic case studies in Part III).
State Space Statistical Arbitrage
Building upon the ML focus from Trading the Unobservable, Triantafyllopoulos and Montana recently published a state space model for statarb spreads entitled Dynamic Modeling of Mean-reverting Spreads for Statistical Arbitrage. This article builds upon the 2005 Elliott et al. article Pairs Trading in JQF.
Regime Discovery
When considering market regime, many traders think first of macro principles: business cycle, risk appetite, interest rates, liquidity, market volatility, etc. This perspective is motivated by applying herd behavior to market dynamics: regimes arise when many people either trade in harmony (trends) or disharmony (range). In this sense, herd behavior is an unobservable for regimes.
Yet, this explanation is not terribly insightful for trading: either one must predict the future or settle for exploiting established trends (i.e. macroeconomic forecasting or trend following). Although both strategies are actively traded, profits tend to be unpredictable.
An alternative approach is trying to build regime discovery models which quantitatively identify regimes early in their formation, and probabilistically inform how to trade ahead of the herd (whether by seconds or days). For example, how price or volatility of HP, VGT, and DJIA change in relation to a price change in IBM. Alternatively, how USD crosses change in response to an unobserved Fed intervention.
Trading the Unobservable
Security prices are driven by diverse factors and constraints, many of which are neither directly observable nor quantifiable by traders. Examples includes fundamental (e.g. corporate actions), behavioral (e.g. herd mentality), financial (e.g. liquidity), macro (e.g. central bank intervention), and microstructure (e.g. market impact algos). Yet, many classic quant models are formulated exclusively using variables which are directly observed: quotes, trades, prices, volumes, spreads, etc.
This is an odd contradiction.
Unraveling this contradiction is central to exploring market regimes, as they defy characterization by observable variables.
Paradox of Informedness
Are the aggregate buy/sell decisions of all traders better than random?
This deceptively simple sounding question has deep, far reaching implications for building quantitative models and high-frequency trading systems. Although many people have speculated on this question for decades, the discipline of econophysics is applying techniques originating from statistical physics (originally statistical mechanics) to provide insight into this fundamental question.
Econophysics frames this question in the context of investor informedness, which is a means of measuring whether the average trade has better than random chance (50/50) to be profitable. Interesting papers in this literature include Fluctuations and Response in Financial Markets by Bouchaud et al. and Are Supply and Demand Driving Stock Prices? by Hopman.
Conclusions to this question drawn from experimental high-frequency order/trade analysis are unexpectedly fascinating, and directly result in a paradox of informedness:
Review: Trading Regime Analysis
The following is a review of Trading Regime Analysis by Gunn, consistent with Quantivity’s on-going market regime theme and in response to reader request.
Readers seeking rigorous quantitative treatment of regime analysis will be disappointed (as frequently and gleefully acknowledged by the author), as the text seeks to analyze regimes primarily through anecdotal behavioral finance and an idiosyncratic basket of technical analysis. That said, the text builds intuition and ultimately endorses numerous intellectual themes which similarly underlie rigorous quantitative analysis of market regimes.
This is admittedly an unexpected curiosity. Thus, value lies in observing how this regime intuition is framed by a certified technical analyst, voyeuristically from the perspective of a quant. Despite TA arguably serving as the intellectual ancestor of quantitative analysis (as introduced in Why Moving Averages), very rarely do technical and modern quantitative analyses converge on intuition. As with any intellectual juxtaposition, such curiosity merits study.
Limit Book Simulation
Michael Kearns and colleagues from Penn and UT did some interesting microstructure work years back with the Penn-Lehman Automated Trading (PLAT) and Penn Exchange Simulator (PXS) projects.
Having read these articles years ago, recent reference by etrading brought back memories.
How to Learn Algorithmic Trading: Part 3
Third in a series on learning quantitative / algorithmic trading, this post focuses on financial modeling and analysis, assuming understanding of financial mathematics from Part 2 and overview of quantitative trading from Part 1. After digesting these, readers should be capable of both building interesting systematic trading systems and understanding microstructure dynamics that drive modern market making (sell side) and large block trading (buy side).
How to Learn Algorithmic Trading: Part 2
Excellent readership and thoughtful comments on the original How to Learn Algorithmic Trading have motivated two follow-up posts on learning quantitative / algorithmic trading (while retrospectively revising the original to improve consistency). This Part focuses on the cross-discipline foundations of financial mathematics, whose knowledge is generally assumed by practitioners and financial modeling literature. The subsequent, Part 3, focuses on modern financial modeling and analysis.
Depending on reader interest, this topic may warrant a future series of posts to delve into seminal literature in selected trading disciplines, such as suggested by etrading on the Penn-Lehman Automated Trading Project.
Quantivity 
