Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/18842
Full metadata record
DC FieldValueLanguage
dc.contributor.authorMittnik, Stefanen_US
dc.contributor.authorZadrozny, Peter A.en_US
dc.date.accessioned2009-01-28T15:53:15Z-
dc.date.available2009-01-28T15:53:15Z-
dc.date.issued2004en_US
dc.identifier.urihttp://hdl.handle.net/10419/18842-
dc.description.abstractThe paper illustrates and evaluates a Kalman filtering method for forecasting German realGDP at monthly intervals. German real GDP is produced at quarterly intervals but analystsand decision makers often want monthly GDP forecasts. Quarterly GDP could be regressedon monthly indicators, which would pick up monthly feedbacks from the indicators to GDP,but would not pick up implicit monthly feedbacks from GDP onto itself or the indicators. Anefficient forecasting model which aims to incorporate all significant correlations in monthlyquarterlydata should include all significant monthly feedbacks. We do this with estimatedVAR(2) models of quarterly GDP and up to three monthly indicator variables, estimatedusing a Kalman-filtering-based maximum-likelihood estimation method. Following themethod, we estimate monthly and quarterly VAR(2) models of quarterly GDP, monthlyindustrial production, and monthly, current and expected, business conditions. The businessconditions variables are produced by the Ifo Institute from its own surveys. We use early insampledata to estimate models and later out-of-sample data to produce and evaluateforecasts. The monthly maximum-likelihood-estimated models produce monthly GDPforecasts. The Kalman filter is used to compute the likelihood in estimation and to produceforecasts. Generally, the monthly German GDP forecasts from 3 to 24 months ahead arecompetitive with quarterly German GDP forecasts for the same time-span ahead, producedusing the same method and the same data in purely quarterly form. However, the presentmixed-frequency method produces monthly GDP forecasts for the first two months of aquarter ahead which are more accurate than one-quarter-ahead GDP forecasts based on thepurely-quarterly data. Moreover, quarterly models based on purely-quarterly data generallycannot be transformed into monthly models which produce equally accurate intra-quarterlymonthly forecasts.en_US
dc.language.isoengen_US
dc.publisher|aCenter for Economic Studies and Ifo Institute (CESifo) |cMunich-
dc.relation.ispartofseries|aCESifo Working Paper |x1203en_US
dc.subject.jelC32en_US
dc.subject.jelE37en_US
dc.subject.ddc330en_US
dc.subject.keywordmixed-frequency dataen_US
dc.subject.keywordVAR modelsen_US
dc.subject.keywordmaximum-likelihood estimationen_US
dc.subject.keywordKalman filteren_US
dc.subject.stwKonjunkturprognoseen_US
dc.subject.stwPrognoseverfahrenen_US
dc.subject.stwVAR-Modellen_US
dc.subject.stwMaximum-Likelihood-Methodeen_US
dc.subject.stwZustandsraummodellen_US
dc.subject.stwSchätzungen_US
dc.subject.stwDeutschlanden_US
dc.titleForecasting quarterly German GDP at monthly intervals using monthly IFO business conditions dataen_US
dc.typeWorking Paperen_US
dc.identifier.ppn85663316Xen_US
dc.rightshttp://www.econstor.eu/dspace/Nutzungsbedingungen-

Files in This Item:
File
Size
412.57 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.