Change Point Estimation in Panel Data without Boundary Issue

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Abstract: Panel data of our interest consist of a moderate number of panels, while the panels contain a small number of observations. An estimator of common breaks in panel means without a boundary issue for this kind of scenario is proposed. In particular, the novel estimator is able to detect a common break point even when the change happens immediately after the first time point or just before the last observation period. Another advantage of the elaborated change point estimator is that it results in the last observation in situations with no structural breaks. The consistency of the change point estimator in panel data is established. The results are illustrated through a simulation study. As a by-product of the developed estimation technique, a theoretical utilization for correlation structure estimation, hypothesis testing and bootstrapping in panel data is demonstrated. A practical application to non-life insurance is presented, as well. Terms of use: Documents in

Common Breaks in Means for Cross‐Correlated Fixed‐T Panel Data

“…By contrast, in panels consistency is usually possible. Bai (2010) was among the first to make this point (see Peštová and Pešta, 2017, for some confirmatory results and a recent overview of the literature). He considered a model in which the mean of the panel data is subject to a common structural break.…”

“…Among the existing fixed-T studies (see again Peštová and Pešta, 2017, for an overview), the one that is closest to our own is that of Horváth et al (2017). They focus on the large-T case; however, they also show that their breakpoint estimator is consistent when T is fixed and that this is true even in the presence of a common factor.…”

“…Random coefficients therefore represent the limit of what we can allow for in terms of cross-section heterogeneity. The random loading assumption is less restrictive than the shrinking condition of Horváth and Hušková (2012), and it is certainly less restrictive than the otherwise so common assumption of at most a common time effect (see, e.g., De Wachter and Tzavalis, 2012;Hidalgo and Schafgans, 2017;Peštová and Pešta, 2017). Random coefficient assumptions are actually very common even in large-T panels, and there is a large and growing literature on factor-augmented regressions with random coefficients (see Chudik and Pesaran, 2015, for a survey).…”

“…E-mail: joakim.westerlund@nek.lu.se 249 passage of time, statistical agencies keep publishing time-series data for individuals, firms, countries, and regions. Thus, while T is usually quite small, N can potentially be very large (see De Wachter and Tzavalis, 2012;Horváth and Hušková, 2012;Li et al, 2016;Hidalgo and Schafgans, 2017;Peštová and Pešta, 2017, for some examples of applications where N is relatively large).…”

Changepoint in dependent and non-stationary panels

“…That means, if there is no change in panel means, the estimator points out at the very last time point T with probability tending to one. Such an estimator was proposed, for instance, by Peštová and Pešta (2017) and it can be easily proved to be consistent also under the above formulated assumptions.…”