Banerjee, Anindya, Robin L. Lumsdaine and James H. Stock, (1992), "Recursive and Sequential Tests of the Unit-Root and Trend-Break Hypotheses: Theory and International Evidence", Journal of Business & Economic Statistics, 10, pp.271-287.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....trends. Structural change in many of these series was confirmed by Vogelsang (1997) and Chu and White (1992) using direct tests for shifts in trend. Vogelsang (1998) found evidence of a mean shift in the unemployment rate. Series for international output were also found to have segmented trends by Banerjee, Lumsdaine and Stock (1992), Ben David and Papell (1995) and Perron (1991) In spite of these findings, many VARs continue to be estimated as though there were no trend breaks in the series. If there are trend breaks in the univariate series, it seems natural that the breaks should also appear in a multivariate system. ....

Banerjee, A., Lumsdaine, R. L. and Stock, J. H. (1992), Recursive and Sequential Tests of the Unit Root and Tread Break Hypothesis, Journal of Business and Economic Statistics 10, 271--287.

....the third author was visiting the Institute of Statistics and Econometrics at the Humboldt University in Berlin. 0 1 Introduction A number of studies consider testing for unit roots in univariate time series which have a level shift. Examples are Perron (1989, 1990) Perron Vogelsang (1992) Banerjee, Lumsdaine Stock (1992), Zivot Andrews (1992) Amsler Lee (1995) Leybourne, Newbold Vougas (1998) Monta n es Reyes (1998) and Saikkonen Lutkepohl (1999) These tests are important because the trending properties of a set of time series determine to some extent which model and statistical procedures are ....

Banerjee, A., R.L. Lumsdaine & J.H. Stock (1992), Recursive and sequential tests of the unit-root and trend-break hypotheses: Theory and international evidence, Journal of Business & Economic Statistics, 10, 271 - 287.

....a structural break has two specific implications for cointegration analysis. First, if a structural break at an priori unknown point in time is present in a time series, the unit root tests proposed by Engle and Granger (1987) have very little power. Better specified test statistics as proposed by Banerjee, Lumsdaine, and Stock (1992) which are consistent even in the presence of structural breaks will therefore need to be employed. Second, a time period free of structural breaks in the cointegration relationship has to be identified otherwise the interpretation of the cointegration vector will result in misleading conclusions. ....

....follow random walks corresponds to H 0 : b=0 for the t statistic and to H 0 : b=x=0 for the F statistic. In the presence of structural breaks at an priory unknown point in time, the above two unit root tests have very little power. Better specified are the following test statistics proposed by Banerjee, Lumsdaine, and Stock (1992). Using equation (1A) and (1B) recursive minimum t statistics can be calculated in order to test the null hypothesis of b=0 3 . Furthermore, sequential unit root tests that distinguish between a shift in the mean or the trend of the series can be calculated based on the following regressions ....

Banerjee, A., R.L. Lumsdaine, and J.H. Stock, 1992, Recursive and sequential test of the unit-root and trendbreak hypotheses: Theory and international evidence, Journal of Business & Economic Statistics 10, 271287.

....rejection of the unit root. For the Leybourne McCabe test, lag selection determines the outcome. SIC does not lead to rejection of trend stationarity (as reported by Cheung and Chinn, 1997) but GS does. 6. The step dummy is highly significant by conventional standards in every case However, Banerjee, Lumsdaine, and Stock (1992) demonstrate that the distributions of break dummy coefficients are non standard. These tests have as maintained hypothesis that the series is homogeneous, generated by an AR process of known order with constant parameters and i.i.d. Normal innovations. It is not clear how deviations from these ....

....AR order. It is the only one of the tests considered here that has too small a size. The size distortion of this test is evidently dependent on the form of the autocorrelation function, since Schwert (1989a) found that the size of this test was too large in the MA(1) case. As noted above, Banerjee, Lumsdaine, and Stock (1992) demonstrate that the asymptotic distributions of break dummy coefficients are nonstandard, although they maximized the F stat of the dummy variable, rather than the unit root statistic, across break dates. Using the tdistribution for inference would create the false impression that a break ....

Banerjee, Anindya, Robin L. Lumsdaine, and James H. Stock, 1992, "Recursive and Sequential Tests of the Unit-Root and Trend-Break Hypotheses: Theory and International Evidence," J. of Business and Economic Statistics, 10, 271-287.

....E 2 and Z 1 statistics, the rejection frequencies show a hump shaped pattern, such that rejection is smallest and largest, respectively, when the break occurs in the middle of the sample. Regarding OE 2 , all of tests display increasingly greater rejection rates as OE 2 decreases, ceteris parabus. Banerjee et al. 1992) found that the ADF test applied to CP processes displayed similar properties to those noted above for the SUR tests. 12 3.8 Structural change in deterministic model part Finally we consider linear AR processes which exhibit structural change in the deterministic part of the model at a certain ....

Banerjee, A., R.L. Lumsdaine and J.H. Stock, 1992, Recursive and sequential tests of the unit-root and trend-break hypotheses: theory and international evidence, Journal of Business & Economic Statistics 10, 271--287.

.... philosophy is to start with the largest available sample and estimate marginal models such as (1b) below, allowing for a single break in the mean and trend (simultaneously) From these equations, by sequential search, we thus first obtain a break date for the full sample using methods developed in Banerjee et al. 1992), Christiano (1992) Zivot and Andrews (1992) and Bai (1997) The sequentiality follows from allowing the location of the break to vary freely across this sample, subject only to trimming restrictions. We then bootstrap these equations n times on the full sample and record the density of the ....

Banerjee, A., R. L. Lumsdaine and J. Stock (1992), "Recursive and Sequential Tests of the Unit-Root and Trend-Break Hypothesis: Theory and International Evidence", Journal of Business and Economic Statistics, 10, 271-288.

....of course that no a priori imposition is undertaken. In our testing methods we operated within the framework of sequential and recursive tests in order to develop reasonably robust and powerful procedures. The distinction between sequential and recursive procedures is described in detail in Banerjee, Lumsdaine and Stock (1992). Recursive procedures involve estimation by considering estimates derived from partial samples and augmenting the sample with additional observations at each stage. Sequential methods use full sample estimates with the location of a particular event being varied across the length of this sample. ....

....later on in the text. Both procedures have trimming restrictions associated with them. We choose to concentrate only on sequential methods for detecting breaks. This decision is based on our experience of recursive tests, confirmed by the simulation analyses in Banerjee and Urga (1995a,b) and in Banerjee, Lumsdaine and Stock (1992), having low power. In contrast with earlier work, we operate wherever necessary within a systems framework and allow for the possibility of distinct structural breaks affecting different points of a system of equation at different points in time. The broader aim of the analysis in this paper is ....

[Article contains additional citation context not shown here]

Banerjee, A., Lumsdaine,R.L. and J.H. Stock (1992), "Recursive and Sequential Tests of the Unit-Root and Trend-Break Hypothesis: Theory and International Evidence," Journal of Business and Economic Statistics, 10(3), 271-250.

No context found.

Banerjee, Anindya, Robin L. Lumsdaine and James H. Stock, (1992), "Recursive and Sequential Tests of the Unit-Root and Trend-Break Hypotheses: Theory and International Evidence", Journal of Business & Economic Statistics, 10, pp.271-287.

No context found.

Banerjee, Anindya, Robin L. Lumsdaine and James H. Stock, (1992), \Recursive and Sequential Tests of the Unit-Root and Trend-Break Hypotheses: Theory and International Evidence", Journal of Business & Economic Statistics, 10, pp.271-287.

No context found.

Banerjee, Anindya, Robin L. Lumsdaine and James H. Stock, (1992), \Recursive and Sequential Tests of the Unit-Root and Trend-Break Hypotheses: Theory and International Evidence", Journal of Business & Economic Statistics, 10, pp.271-287.

No context found.

Banerjee, A., Lumsdaine, R., and J. Stock (1992). "Recursive and Sequential Tests of the Unit Root and Trend Break Hypotheses: Theory and International Evidence," Journal of Business and Economic Statistics, v. 10, no. 3, 271-288.

No context found.

Banerjee, A., R. Lumsdaine, and J.H. Stock (1992) "Recursive and Sequential Tests of the Unit Root and Trend Break Hypotheses: Theory and International Evidence", Journal of Business and Economic Statistics, 10, 271-288.

No context found.

Banerjee, A., R. L. Lumsdaine and J. H. Stock, 1992, \Recursive and Sequential Tests of the Unit root and Trend Break Hypothesis: Theory and International Evidence," Journal of business and Economic Statistics, 10, 271-287.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC