Testing for a break in persistence under long‐range dependencies

Abstract

Keywords: break in pesistence, long memory, CUSUM of squares based test.

Long memory and changing persistence

“…More details on the asymptotic distribution of R and critical values can be found in Sibbertsen and Kruse (2009). The simulation results therein show that the test is correctly sized for the sample sizes considered here.…”

“…To this end, we study the behaviour of a CUSUM of squares-based test suggested by Sibbertsen and Kruse (2009 …”

“…A leading example are inflation rates which are modeled by either (i) changing persistence (Halunga et al 2008 andNoriega andRamos-Francia 2009) or (ii) long memory models (see Hassler and Wolters 1995, Hsu 2005and Lee 2005. In order to discriminate between long memory and changing persistence, or stable shifts, we suggest to use a CUSUM of squares-based test proposed by Sibbertsen and Kruse (2009). Further simulation results show that it has remarkable power to detect spuriously generated long memory due to structural changes in the autoregressive parameters.…”

The Memory of Stock Return Volatility: Asset Pricing Implications

“…Leybourne et al (2007) consider long memory dynamics and introduce a test for a break from stationary long memory to non-stationary long memory. Their test is improved by Sibbertsen & Kruse (2009), since the results may be distorted when the data-generating process exhibits long memory. They apply the test to U.S. inflation data and find a break in the early 1980s.…”

Measuring stationarity in long-memory processes

“…In fact, although stationary long-memory models are often employed in practice, there do not exist many tests for the hypothesis of stationarity which include these processes. Berkes et al (2006), Sibbertsen andKruse (2009) andDehling, Rooch, andTaqqu (2013) consider CUSUM and Wilcoxon type tests to discriminate between long-range dependence and one change in mean. A change with respect to the mean is of course only the simplest possible deviation from stationarity and there is particular interest in measuring deviations in the dependency structure over time as well.…”