Changepoint estimation for dependent and non-stationary panels

Michal Pešta; Barbora Peštová; Matúš Maciak

Applications of Mathematics (2020)

  • Volume: 65, Issue: 3, page 299-310
  • ISSN: 0862-7940

Abstract

top
The changepoint estimation problem of a common change in panel means for a very general panel data structure is considered. The observations within each panel are allowed to be generally dependent and non-stationary. Simultaneously, the panels are weakly dependent and non-stationary among each other. The follow up period can be extremely short and the changepoint magnitudes may differ across the panels accounting also for a specific situation that some magnitudes are equal to zero (thus, no jump is present in such case). We introduce a novel changepoint estimator without a boundary issue meaning that it can estimate the change close to the extremities of the studied time interval. The consistency of the nuisance-parameter-free estimator is proved regardless of the presence/absence of the change in panel means under relatively simple conditions. Empirical properties of the proposed estimator are investigated through a simulation study.

How to cite

top

Pešta, Michal, Peštová, Barbora, and Maciak, Matúš. "Changepoint estimation for dependent and non-stationary panels." Applications of Mathematics 65.3 (2020): 299-310. <http://eudml.org/doc/297130>.

@article{Pešta2020,
abstract = {The changepoint estimation problem of a common change in panel means for a very general panel data structure is considered. The observations within each panel are allowed to be generally dependent and non-stationary. Simultaneously, the panels are weakly dependent and non-stationary among each other. The follow up period can be extremely short and the changepoint magnitudes may differ across the panels accounting also for a specific situation that some magnitudes are equal to zero (thus, no jump is present in such case). We introduce a novel changepoint estimator without a boundary issue meaning that it can estimate the change close to the extremities of the studied time interval. The consistency of the nuisance-parameter-free estimator is proved regardless of the presence/absence of the change in panel means under relatively simple conditions. Empirical properties of the proposed estimator are investigated through a simulation study.},
author = {Pešta, Michal, Peštová, Barbora, Maciak, Matúš},
journal = {Applications of Mathematics},
keywords = {panel data; changepoint; change in means; estimation; dependence; non-stationarity; call options; non-life insurance},
language = {eng},
number = {3},
pages = {299-310},
publisher = {Institute of Mathematics, Academy of Sciences of the Czech Republic},
title = {Changepoint estimation for dependent and non-stationary panels},
url = {http://eudml.org/doc/297130},
volume = {65},
year = {2020},
}

TY - JOUR
AU - Pešta, Michal
AU - Peštová, Barbora
AU - Maciak, Matúš
TI - Changepoint estimation for dependent and non-stationary panels
JO - Applications of Mathematics
PY - 2020
PB - Institute of Mathematics, Academy of Sciences of the Czech Republic
VL - 65
IS - 3
SP - 299
EP - 310
AB - The changepoint estimation problem of a common change in panel means for a very general panel data structure is considered. The observations within each panel are allowed to be generally dependent and non-stationary. Simultaneously, the panels are weakly dependent and non-stationary among each other. The follow up period can be extremely short and the changepoint magnitudes may differ across the panels accounting also for a specific situation that some magnitudes are equal to zero (thus, no jump is present in such case). We introduce a novel changepoint estimator without a boundary issue meaning that it can estimate the change close to the extremities of the studied time interval. The consistency of the nuisance-parameter-free estimator is proved regardless of the presence/absence of the change in panel means under relatively simple conditions. Empirical properties of the proposed estimator are investigated through a simulation study.
LA - eng
KW - panel data; changepoint; change in means; estimation; dependence; non-stationarity; call options; non-life insurance
UR - http://eudml.org/doc/297130
ER -

References

top
  1. Bai, J., 10.1016/j.jeconom.2009.10.020, J. Econom. 157 (2010), 78-92. (2010) Zbl06608388MR2652280DOI10.1016/j.jeconom.2009.10.020
  2. Baltagi, B. H., Feng, Q., Kao, C., 10.1016/j.jeconom.2015.03.048, J. Econom. 191 (2016), 176-195. (2016) Zbl1390.91250MR3434442DOI10.1016/j.jeconom.2015.03.048
  3. Bradley, R. C., 10.1214/154957805100000104, Probab. Surv. 2 (1986), 107-144. (1986) Zbl1189.60077MR2178042DOI10.1214/154957805100000104
  4. Wachter, S. De, Tzavalis, E., 10.1016/j.csda.2012.02.025, Comput. Stat. Data Anal. 56 (2012), 3020-3034. (2012) Zbl1254.91002MR2943878DOI10.1016/j.csda.2012.02.025
  5. Horváth, L., Hušková, M., 10.1111/j.1467-9892.2012.00796.x, J. Time Ser. Anal. 33 (2012), 631-648. (2012) Zbl1282.62181MR2944843DOI10.1111/j.1467-9892.2012.00796.x
  6. Kim, D., 10.1016/j.jeconom.2011.06.018, J. Econom. 164 (2011), 310-330. (2011) Zbl06616780MR2826773DOI10.1016/j.jeconom.2011.06.018
  7. Lin, Z., Lu, C., Limit Theory for Mixing Dependent Random Variables, Mathematics and Its Applications 378. Kluwer Academic Publishers, Dordrecht; Science Press Beijing, New York (1996). (1996) Zbl0889.60001MR1486580
  8. Maciak, M., 10.1016/j.ecosta.2019.12.005, Econom. Stat. (2020), 10 pages. (2020) DOI10.1016/j.ecosta.2019.12.005
  9. Maciak, M., Pešta, M., Peštová, B., 10.1007/s00362-020-01180-6, (to appear) in Stat. Pap. (2020). DOI10.1007/s00362-020-01180-6
  10. Pesaran, M. H., 10.1111/j.1468-0262.2006.00692.x, Econometrica 74 (2006), 967-1012. (2006) Zbl1152.91718MR2238209DOI10.1111/j.1468-0262.2006.00692.x
  11. Pešta, M., 10.1080/02331888.2012.658806, Statistics 47 (2013), 966-991. (2013) Zbl06231528MR3175728DOI10.1080/02331888.2012.658806
  12. Peštová, B., Pešta, M., 10.1007/s00184-014-0522-8, Metrika 78 (2015), 665-689 Erratum (2016), 237-238. (2015) Zbl1333.62154MR3369338DOI10.1007/s00184-014-0522-8
  13. Peštová, B., Pešta, M., 10.3390/risks5010007, Risks 5 (2017), Article ID 7, 22 pages. (2017) DOI10.3390/risks5010007

NotesEmbed ?

top

You must be logged in to post comments.