Common Nonstationary Components of Asset Prices
ABSTRACT This paper presents a stock-flow consistent macroeconomic model in which financial fragility in firm and household sectors evolves endogenously through the interaction between real and financial sectors. Changes in firms' and households' financial practices produce long waves. The Hopf bifurcation theorem is applied to clarify the conditions for the existence of limit cycles, and simulations illustrate stable limit cycles. The long waves are characterized by periodic economic crises following long expansions. Short cycles, generated by the interaction between effective demand and labor market dynamics, fluctuate around the long waves.
- SourceAvailable from: sugiyama-www.cs.titech.ac.jp
- "Canonical correlation analysis (CCA) (Hotelling, 1936) is a classical dimensionality reduction technique for two data sources, which finds projection directions with maximum correlation. CCA has been successfully applied in various fields such as neuroscience (Becker, 1996; Becker & Hinton, 1992; Favorov & Ryder, 2004), econometrics (Bossaerts, 1988; Vinod, 1968), psychometrics (McKeon, 1967), meteorology (Storch & Zwiers, 2002), bioinformatics (Gumus et al., 2012; Naylor et al., 2010; Vert & Kanehisa, 2003; Yamanishi et al., 2003), and information retrieval (Hardoon et al., 2004; Li & Shawe-Taylor, 2006; Vinokourov et al., 2003). Although CCA has been originally developed as an unsupervised learning method, it is also closely related to supervised tasks such as classification. "
[Show abstract] [Hide abstract]
- "In its most common form, pairs trading involves forming a portfolio of two related stocks whose relative pricing is away from its "equilibrium" state. It is linked to co-integration (Bossaerts, 1988) and correlation in stock prices, mean reversion (DeBondt and Thaler, 1985), contrarian strategies (Jegadeesh and Titman, 1993) and also to the law of the one price. Pairs trading is one way to select and build stocks for a long/short dollar neutral portfolio. "
ABSTRACT: Pairs trading is a popular speculation strategy. Several implementation methods are proposed in the literature: they can be based on a distance criterion or on co-integration. This article extends previous research in another direction: the combination of forecasting techniques (Neural Networks) and multi-criteria decision making methods (Electre III). The key contribution of this paper is the introduction of multi-step-ahead forecasts. It leads to major changes in the trading system and raises new empirical and methodological questions. The results of an application based on S&P 100 Index stocks are promising: this methodology could be a powerful tool for pairs selection in a highly non-linear environment.European Journal of Operational Research 12/2010; 207(3-207):1702-1716. DOI:10.1016/j.ejor.2010.06.043 · 2.36 Impact Factor
[Show abstract] [Hide abstract]
- "Furthermore, there are several works available that employ cointegration theory to other fields of research. Bossaerts (1988), for instance, developed a test of cointegration and applied it to size-based and industry-based stock portfolios. His results provide substantial evidence of cointegration. "
ABSTRACT: An advantageous statistical arbitrage strategy should exhibit a zero-cost trading strategy for which the expected payoff should be positive. In practical applications, however, the abnormal returns often are out-of-sample not significant. The statistical model being suggested here results in an estimated portfolio exhibiting in-sample a cointegration relationship with the artificial stock index. The portfolio returns exhibited out-of-sample a mean of 10.44% p.a., whereas the volatility was one third lower in comparison to the benchmark's volatility. Accounting for trading costs of 2.94% p.a. on average, the annual returns of the estimated portfolio are out-of-sample still 6.83% higher than the market returns. As a result, the model involves implicitly advantageous market timing.