Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Replication, development and evaluation of a GDP indicator for the Swedish business cycle. Can the framework of ECB’s indicator ALI be used to create an indicator for the Swedish business cycle?

Svensson, Elina LU (2014) NEKN01 20141
Department of Economics
Abstract
This study shows that the framework used to construct the successful Euro Area Leading Indicator (ALI), developed by de Bondt and Hahn on behalf of the European Central Bank, can successfully be applied to the Swedish business cycle. The ALI type indicator constructed in this study is a weighted index of nine macroeconomic series and accurately predicts the Swedish business cycle by three months (or longer). A data driven indicator using Principal Component Analysis is constructed for comparison, and leads the Swedish business cycle by seven months. It has zero bias whereas the ALI Sweden indicator slightly underestimates the business cycle, but is slightly more accurate than the data driven indicator. Both indicators are efficient. An... (More)
This study shows that the framework used to construct the successful Euro Area Leading Indicator (ALI), developed by de Bondt and Hahn on behalf of the European Central Bank, can successfully be applied to the Swedish business cycle. The ALI type indicator constructed in this study is a weighted index of nine macroeconomic series and accurately predicts the Swedish business cycle by three months (or longer). A data driven indicator using Principal Component Analysis is constructed for comparison, and leads the Swedish business cycle by seven months. It has zero bias whereas the ALI Sweden indicator slightly underestimates the business cycle, but is slightly more accurate than the data driven indicator. Both indicators are efficient. An out-of-sample real time evaluation confirms the indicators’ good performance. A comparison of the series filtered with the HP-filter and the one sided RW-filter shows that the RW-filtered is superior as it filters out the high frequent noise and doesn’t introduce spurious cycles. Service Production Index is found to not be a better proxy for the business cycle than Industrial Production Index. This study suggests the use of both indicators: the ALI Sweden indicator for its accurate and transparent forecast and the data driven indicator for its long lead time and zero bias. (Less)
Please use this url to cite or link to this publication:
author
Svensson, Elina LU
supervisor
organization
course
NEKN01 20141
year
type
H1 - Master's Degree (One Year)
subject
keywords
leading indicator, index, Principal Component Analysis, data-driven, turning point, filter, bias, efficiency, accuracy
language
English
id
4616640
date added to LUP
2014-09-25 13:51:36
date last changed
2014-09-25 13:51:36
@misc{4616640,
  abstract     = {{This study shows that the framework used to construct the successful Euro Area Leading Indicator (ALI), developed by de Bondt and Hahn on behalf of the European Central Bank, can successfully be applied to the Swedish business cycle. The ALI type indicator constructed in this study is a weighted index of nine macroeconomic series and accurately predicts the Swedish business cycle by three months (or longer). A data driven indicator using Principal Component Analysis is constructed for comparison, and leads the Swedish business cycle by seven months. It has zero bias whereas the ALI Sweden indicator slightly underestimates the business cycle, but is slightly more accurate than the data driven indicator. Both indicators are efficient. An out-of-sample real time evaluation confirms the indicators’ good performance. A comparison of the series filtered with the HP-filter and the one sided RW-filter shows that the RW-filtered is superior as it filters out the high frequent noise and doesn’t introduce spurious cycles. Service Production Index is found to not be a better proxy for the business cycle than Industrial Production Index. This study suggests the use of both indicators: the ALI Sweden indicator for its accurate and transparent forecast and the data driven indicator for its long lead time and zero bias.}},
  author       = {{Svensson, Elina}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Replication, development and evaluation of a GDP indicator for the Swedish business cycle. Can the framework of ECB’s indicator ALI be used to create an indicator for the Swedish business cycle?}},
  year         = {{2014}},
}