Advanced

Is there hysteresis in unemployment rates?

Lavesson, Niclas LU (2012) STAH11 20112
Department of Statistics
Abstract
This paper investigates if there is hysteresis in unemployment rates for 8 OECD countries. Univariate unit root tests (ADF and KPSS tests) are performed as an initial analysis of the unemployment rates. These tests suggest that the hysteresis hypothesis is true.

Since traditional unit root test have low power against fractionally integrated alternatives, the Sowell’s (1992) exact maximum likelihood (EML) ARFIMA estimator is used in the paper as well. The main advantage of using this estimator is since a quantitative measure of the memory parameters is given. As a consequence, tests of the estimated memory parameters can be conducted and a more refined analysis is possible. When the EML estimator is used, the result found by the ADF and... (More)
This paper investigates if there is hysteresis in unemployment rates for 8 OECD countries. Univariate unit root tests (ADF and KPSS tests) are performed as an initial analysis of the unemployment rates. These tests suggest that the hysteresis hypothesis is true.

Since traditional unit root test have low power against fractionally integrated alternatives, the Sowell’s (1992) exact maximum likelihood (EML) ARFIMA estimator is used in the paper as well. The main advantage of using this estimator is since a quantitative measure of the memory parameters is given. As a consequence, tests of the estimated memory parameters can be conducted and a more refined analysis is possible. When the EML estimator is used, the result found by the ADF and KPSS tests are strengthened; there is hysteresis in unemployment rates. (Less)
Please use this url to cite or link to this publication:
author
Lavesson, Niclas LU
supervisor
organization
course
STAH11 20112
year
type
M2 - Bachelor Degree
subject
keywords
Hysteresis, unemployment, ARFIMA, fractional integration, time series
language
English
id
2371761
date added to LUP
2012-04-02 20:04:25
date last changed
2012-04-02 20:04:25
@misc{2371761,
  abstract     = {This paper investigates if there is hysteresis in unemployment rates for 8 OECD countries. Univariate unit root tests (ADF and KPSS tests) are performed as an initial analysis of the unemployment rates. These tests suggest that the hysteresis hypothesis is true.

Since traditional unit root test have low power against fractionally integrated alternatives, the Sowell’s (1992) exact maximum likelihood (EML) ARFIMA estimator is used in the paper as well. The main advantage of using this estimator is since a quantitative measure of the memory parameters is given. As a consequence, tests of the estimated memory parameters can be conducted and a more refined analysis is possible. When the EML estimator is used, the result found by the ADF and KPSS tests are strengthened; there is hysteresis in unemployment rates.},
  author       = {Lavesson, Niclas},
  keyword      = {Hysteresis,unemployment,ARFIMA,fractional integration,time series},
  language     = {eng},
  note         = {Student Paper},
  title        = {Is there hysteresis in unemployment rates?},
  year         = {2012},
}