Data & Analytics

The Seasonal Adjustment of the Consumer and Wholesale Prices: a Comparison of Census X-11, X-12 ARIMA and TRAMO/SEATS

The Seasonal Adjustment of the Consumer and Wholesale Prices: a Comparison of Census X-11, X-12 ARIMA and TRAMO/SEATS
of 35
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Research Department Working Paper No:  5 The Seasonal Adjustment of the Consumerand Wholesale Prices: a Comparison of CensusX-11, X-12 ARIMA and TRAMO/SEATSJune 2002 The Central Bank of the Republic of Turkey   Preliminary Draft Preliminary Draft Preliminary Draft Preliminary Draft  The Seasonal Adjustment of the Consumerand Wholesale Prices : a Comparison of CensusX-11, X-12 ARIMA and TRAMO/SEATS *   by Meltem Gulenay Ongan  ♣   June 2002  Abstract Abstract Abstract Abstract The paper aims at analyzing the seasonal movements of consumerprices and wholesale prices with respect to different seasonal adjustmenttechniques. Suitability of different seasonal adjustment techniques will betested for CPI, WPI, CPI excluding food prices, and WPI excludingagricultural prices. The literature on the analysis of seasonality iscomposed of different groups. One group treats the seasonality as a noisethat contaminates the economic data. Another group treats seasonality asa more integrated part of the modeling strategy. However, the most well-known method among the seasonal adjustment procedure is the treatmentto the problem of seasonality as a noise, and this is the method that hasbeen widely applied by the statistical offices of different countries to createofficial seasonally adjusted data. In this respect, the most commonlyapplied official seasonal adjustment procedure has for many years beenthe X-11 method developed at the US Bureau of the Census. The X-11procedure has been replaced by X-12. Among the two methodology, in somecountries and in the EU statistical office EUROSTAT, a programTRAMO/SEATS is applied as well in recent years. After comparing thesedifferent procedures, the paper concludes with the analysis of whichseasonal adjustment procedure may better fit for the Turkish CPI and WPIindices.KeywordsKeywordsKeywordsKeywords: seasonality, Census X-11, X-12 ARIMA, TRAMO/SEATSJel ClassificationJel ClassificationJel ClassificationJel Classification:C20, E31. * The views expressed in this paper are those of the author and do not necessarily reflect theviews of the Cenral Bank of the Republic of Turkey. ♣ The Central Bank of the Republic of Turkey, Research Department,  2    Acknowledgements Acknowledgements Acknowledgements Acknowledgements I would like to thank the CBRT Research Department for giving methe opportunity and providing me with a nice environment for carrying outsuch a study. I would like to thank Zafer Yukseler for his encouragementto work on the seasonal adjustment issue, Devrim Yavuz for thepreparation of choosing a the base year part. Finally I want to express myspecial thanks to Burcu Eyigungor whose comments and edits hadvaluable contributions in this study.  3 1. Introduction1. Introduction1. Introduction1. Introduction The most important motivation behind the study of seasonaladjustment is the lack of an institutionally determined seasonallyadjusted time series for Turkey, especially prior to the inflationtargeting framework, during which the seasonally adjusted priceseries and the seasonal components of the series in question will besignificant for the monetary policy. In this framework, this paper aimsat forming a small study for Turkish price series with respect to theinformation about the different seasonal adjustment procedures thathas been implemented in different statistics offices of differentcountries and to examine the characteristics of Turkey’s consumer andwholesale price series using these different adjustment procedures.The issue of seasonal adjustment is very complicated as it is atechnical issue that has a considerable policy-wise conclusions. To givean example about how the seasonal adjustment issue is importantboth on the technical and policy formation side, one can refer toMaravall 1996:“ For example, what are the standard errors associated withthe estimated seasonal factors? This point has important  policy implications. In short-term monetary control, if the monthly target for the rate of growth of M1 (seasonally adjusted) is 10% and actual growth for that month turns out to be 13%, can we conclude that growth has been excessive and raise, as a consequence, short-term interest rates? Can the 3 percent points difference be attributed to the error implied by the estimation of the seasonally adjusted series? Similarly when assessing the evolution of unemployment, if  the series of total employment grows by 90.000 persons in quarter, and the seasonal effect for that quarter is estimated as an increase of 50.000 persons, can we assume that the increase has been more than a pure seasonal effect? (…)”  In line with what Maravall considered in his study, the sameissue can be implemented to the Turkish price data, especially withinthe first half of 2002 in which the realizations for the agricultural andfood prices are beyond the pure seasonal effect. When considering theseasonally adjustment issue, especially for a country that has a highlyvolatile price data that has a high percentage of outliers, one may bebounded with many questions related with the most suitableadjustment procedure that fits the time-series data, which hasunstable characteristics. In this respect, the paper will try to raisesome important questions which may be a subject for other studies inthe area of seasonal adjustment for Turkey’s price indices and try to  4 seek for possible solutions to the questions that are raised related withthe seasonal adjustment issue.One of the questions that will be raised within the context of thispaper is choosing a starting year of seasonal adjustment that will bediscussed in the third part of the paper. In the fourth part, differentadjustment methods within the framework of their historical evolutionand their properties will be scrutinized in the fourth part of the paper.Finally, the price series analyzed within the paper will be consideredwith respect to suitability to seasonal adjustment, and how theseseries can be used for short-term monitoring and forecasting of inflation. The paper concludes with the importance of seasonaladjustment with respect to central bank policy. As a matter of fact,besides price indices, seasonal adjustment of many other indices areimportant in terms of the central bank policy, like capacity utilizationrate, gross domestic product and unemployment rate because thecurrent business condition of the economy does have an effect on thefuture inflation rate. Nevertheless, in this study, the focus will be onthe deseasonalization of different price indices, the CPI and WPI, andtheir sub-items. 2. Seasonal Adjustment and Characteristics of the2. Seasonal Adjustment and Characteristics of the2. Seasonal Adjustment and Characteristics of the2. Seasonal Adjustment and Characteristics of theTurkey’s Price IndicesTurkey’s Price IndicesTurkey’s Price IndicesTurkey’s Price Indices For the purposes of economic analysis and to set short-termmacroeconomic policy, governments frequently use seasonally adjustedseries, on occasion, trends. Statistical offices generally publishinformation on the national economies using seasonally adjustedseries. Seasonal adjustment has a special importance, especially withrespect to inflation as the turning points of inflation is crucial for theconduct of monetary policy.The aim of the seasonal analysis is to isolate the seasonalfluctuations, generally with the method of decomposing time-seriesinto its different components. These components are usually selectedas a trend component, a cyclical component, a seasonal component, anirregular or random component and lastly a trend break.The trend component of a time-series shows the general tendencyof the non-recurring movement of the series over a long period of time.The cyclical component includes fluctuations over 3 to 11 years thatare related with the short waves. 1   1 Early investigators found that it was not possible to uniquely decompose the trend andcycle components. Thus, these were grouped together; the resulting component isusually referred to as the “trend cycle component”. This component gathers two parts: along-term trend from general phenomena of growth or decrease usually linked to
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!