In a companion paper, Perron and Zhou (2008) provided a comprehensive treatment of the problem of testing jointly for structural change in both the regression coefficients and the variance of the errors in a single equation regression model involving stationary regressors, allowing the break dates for the two components to be different or overlap. The aim of this paper is twofold. First, we pre...

Using data from Germany, Japan, UK, and the U.S., we explore possible threshold cointegration in nominal shortand long-run interest rates with corresponding inflation rates. Traditional cointegration implies perfect mean reversion in real rates and hence confirms the Fisher hypothesis. Threshold cointegration accounts for the possibility that this mean reversion is active only conditional on ce...

This paper explores the implications of asset return predictability on longterm portfolio choice when return forecasting variables exhibit long memory. We model long memory using the class of fractionally integrated time series models. Important predictor variables for U.S. data, like the dividend-price ratio and nominal and real interest rates, are non-stationary with orders of integration aro...

We investigate the stationarity of daily real stock prices in 12 Asia-Pacific countries over period 1991–2020. The methodology employed is driven by need to address three key concerns: (i) identification association between size shocks and stationarity; (ii) different speeds adjustment towards long-run equilibrium; (iii) mean reversion potential asymmetric speed before after 2008–2009 global fi...

In this paper, we present a new method for modeling timeevolving correlation networks, using a Mean Reversion Autoregressive Model, and apply this to stock market data. The work is motivated by the assumption that the price and return of a stock eventually regresses back towards their mean or average. This allows us to model the stock correlation time-series as an autoregressive process with a ...

We study the drift of stationary diffusion processes in a time series analysis of the autoregression function. A marked empirical process measures the difference between the nonparametric regression functions of two time series. We bootstrap the distribution of a Kolmogorov-Smirnov-type test statistic for two hypotheses: Equality of regression functions and shifted regression functions. Neither...

Given multivariate time series, we study the problem of forming portfolios with maximum mean reversion while constraining the number of assets in these portfolios. We show that it can be formulated as a sparse canonical correlation analysis and study various algorithms to solve the corresponding sparse generalized eigenvalue problems. After discussing penalized parameter estimation procedures, ...

a r t i c l e i n f o This study examines whether mean reversion in REIT prices presents an asymmetric behavior across various quantiles. Distinguished from previous literature that applied the traditional linear unit-root test, a state-of-the-art quantile unit-root test is employed to identify financial asset predictability in five real estate investment trust (REIT) classifications. Our empir...

Using a data set on more than 300 UK pension funds' asset holdings, this paper provides a systematic investigation of the performance of managed portfolios across multiple asset classes. We ̄nd evidence of slow mean reversion in the funds' portfolio weights towards a common, time-varying strategic asset allocation. We also ̄nd surprisingly little cross-sectional variation in the average ex post...