Skip to main content
Log in

Forecasting gross value-added at the regional level: are sectoral disaggregated predictions superior to direct ones?

  • Original Paper
  • Published:
Review of Regional Research Aims and scope Submit manuscript


In this paper, we ask whether it is possible to forecast gross value-added (GVA) and its sectoral sub-components at the regional level. With an autoregressive distributed lag model we forecast total and sectoral GVA for one German state (Saxony) with more than 300 indicators from different regional levels (international, national and regional) and additionally make usage of several forecast pooling strategies and factor models. Our results show that we are able to increase forecast accuracy of GVA for every sector and for all forecast horizons (one up to four quarters) compared to an autoregressive process. Finally, we show that sectoral forecasts contain more information in the short term (one quarter), whereas direct forecasts of total GVA are preferable in the medium (two and three quarters) and long term (four quarters).

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save

Springer+ Basic
EUR 32.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or Ebook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2

Similar content being viewed by others


  1. Germany consists of 16 different states which are categorized as NUTS 1 for statistics of the European Union. In comparison, Germany is classified as NUTS 0.

  2. Eastern Germany is the aggregation of five German states: Brandenburg, Mecklenburg-West Pomerania, the Free State of Saxony, Saxony-Anhalt and the Free State of Thuringia.

  3. The data are available upon request from

  4. These six sectors describe the whole economy so that the sum of these sectors equals total GVA.

  5. Drechsel and Scheufele (2012a) state that in most cases simple averages are used for weighting sub-components. In contrast, they use a moving average over the last four quarters to obtain their estimated weights. Since the shares in our sample are relatively persistent, the results should not differ dramatically by applying another approach.

  6. In this paper we denote one quarter (h=1) as short term, two and three quarters (h=2,3) as medium term and four quarters (h=4) as long term. These definitions are in line with the forecasting literature and do not reflect time horizons in macroeconomic theory.

  7. The literature has not found a consesus yet about the level of the discount rate. We apply different values (δ∈{0,0.1,0.2,…,1}) and find similar results. Because of this and to avoid long tables, we only report the outcome for a discount rate equal to 0.1.

  8. We do not take into account the ragged edge problem (see Wallis, 1986) and extract the factors from the information set up to t−1.

  9. To understand the notation in the results section, the following example should make it clear. Imagine a factor model is abbreviated with QML1QSOLS. Then one common factor (1) is extracted via quasi-maximum likelihood (QML) from quarterly data (Q) and the forecast is generated from an OLS estimation. In this case, the factors are obtained from the set of Saxon indicators (S).

  10. Detailed results for all sectors are available upon request.

  11. See Abberger and Wohlrabe (2006) for a recent survey for Germany. For an analysis for Saxony, see Lehmann et al. (2010).


  • Abberger K, Wohlrabe K (2006) Einige Prognoseeigenschaften des ifo Geschäftsklimas – Ein Überblick über die neuere wissenschaftliche Literatur. Ifo-Schnelld 59(22):19–26

    Google Scholar 

  • Angelini E, Banbura M, Rünstler G (2010) Estimating and forecasting the euro area monthly national accounts from a dynamic factor model. J Bus Cycle Meas Anal 2010(1):5–26

    Google Scholar 

  • Bandholz H, Funke M (2003) Die Konstruktion und Schätzung eines Konjunkturfrühindikators für Hamburg. Wirtschaftsdienst 83(8):540–548

    Google Scholar 

  • Barhoumi KP, Brunhes-Lesage V, Darné O, Ferrara L, Pluyaud B, Rouvreau B (2008) Monthly forecasting of French GDP: a revised version of the OPTIM model. Banque de France working papers No 222

  • Barhoumi KP, Darné O, Ferrara L, Pluyaud B (2012) Monthly GDP forecasting using bridge models: application for the French economy. Bull Econ Res 64:s1:s53–s70

    Article  Google Scholar 

  • Chow GC, Lin A (1971) Best linear unbiased interpolation, distribution and exploration of time series by related series. Rev Econ Stat 53(4):372–375

    Article  Google Scholar 

  • Clements HP, Hendry DF (1993) On the limitations of comparing mean squared forecast errors. J Forecast 12(8):617–637

    Article  Google Scholar 

  • Cors A, Kuzin V (2003) An approach for timely estimations of the German GDP. AStA Adv Stat Anal 87(2):201–220

    Google Scholar 

  • Diebold FX, Mariano RS (1995) Comparing predictive accuracy. J Bus Econ Stat 13(3):253–263

    Google Scholar 

  • Doz C, Giannone D, Reichlin L (2011) A two-step estimator for large approximate dynamic factor models based on Kalman filtering. J Econom 164(1):188–205

    Article  Google Scholar 

  • Doz C, Giannone D, Reichlin L (2012) A quasi-maximum likelihood approach for large, approximate dynamic factor models. Rev Econ Stat 94(4):1014–1024

    Article  Google Scholar 

  • Drechsel K, Maurin L (2011) Flow of conjunctural information and forecast of euro area economic activity. J Forecast 30(3):336–354

    Article  Google Scholar 

  • Drechsel K, Scheufele R (2012a) Bottom-up or direct? Forecasting German GDP in a data-rich environment. Swiss national bank working papers 2012-16

  • Drechsel K, Scheufele R (2012b) The performance of short-term forecasts of the German economy before and during the 2008/2009 recession. Int J Forecast 28(2):428–445

    Article  Google Scholar 

  • Dreger C, Kholodilin KA (2007) Prognosen der regionalen Konjunkturentwicklung. Q J Econ Res 76(4):47–55

    Google Scholar 

  • Eickmeier S, Ziegler C (2008) How successful are dynamic factor models at forecasting output and inflation? A meta-analytic approach. J Forecast 27(3):237–265

    Article  Google Scholar 

  • Forni M, Halli M, Lippi M, Reichlin L (2005) The generalized dynamic factor model: one-sided estimation and forecasting. J Am Stat Assoc 100(471):830–840

    Article  Google Scholar 

  • German Council of Economic Experts (2008) Das deutsche Finanzsystem: Effizienz steigern – Stabilität erhöhen. Expertise commissioned by the Federal Government. German Council of Economic Experts

  • Giannone D, Reichlin L, Small D (2008) Nowcasting: the real-time informational content of macroeconomic data. J Monet Econ 55(4):665–676

    Article  Google Scholar 

  • Granger CWJ, Newbold P (1973) Some comments on the evaluation of economic forecasts. Appl Econ 5(1):35–47

    Article  Google Scholar 

  • Hahn E, Skudelny F (2008) Early estimates of euro area real GDP growth—a bottom up approach from the production side. ECB working paper series No 975

  • Harvey DI, Leybourne SJ, Newbold P (1997) Testing the equality of prediction mean squared errors. Int J Forecast 13(2):281–291

    Article  Google Scholar 

  • Harvey DI, Leybourne SJ, Newbold P (1998) Tests for forecast encompassing. J Bus Econ Stat 16(2):254–259

    Google Scholar 

  • Kholodilin KA, Kooths S, Siliverstovs B (2008) A dynamic panel data approach to the forecasting of the GDP of German Länder. Spatial Econ Anal 3(2):195–207

    Article  Google Scholar 

  • Kopoin A, Moran K, Paré JP (2013) Forecasting regional GDP with factor models: how useful are national and international data? Econ Lett 121(2):267–270

    Article  Google Scholar 

  • Lehmann R, Wohlrabe K (2013) Forecasting GDP at the regional level with many predictors. Ger Econ Rev. doi:10/1111/geer.12042

    Google Scholar 

  • Lehmann R, Speich WD, Straube R, Vogt G (2010) Funktioniert der ifo Konjunkturtest auch in wirtschaftlichen Krisenzeiten? Eine Analyse der Zusammenhänge zwischen ifo Geschäftsklima und amtlichen Konjunkturdaten für Sachsen. ifo Dresd ber 17(3):8–14

    Google Scholar 

  • Marcellino M, Stock JH, Watson MW (2003) Macroeconomic forecasting in the euro area: country specific versus area-wide information. Eur Econ Rev 47(1):1–18

    Article  Google Scholar 

  • Newey WK, West KD (1987) A simple positive semi-definite heteroskedasticity and autocorrelation consistent covariance matrix. Econom J 55(3):703–708

    Google Scholar 

  • Nierhaus W (2007) Vierteljährliche Volkswirtschaftliche Gesamtrechnungen für Sachsen mit Hilfe temporaler Disaggregation. ifo Dresd Ber 14(4):24–36

    Google Scholar 

  • Schumacher C (2007) Forecasting German GDP using alternative factor models based on large datasets. J Forecast 26(4):271–302

    Article  Google Scholar 

  • Schumacher C (2010) Factor forecasting using international targeted predictors: the case of German GDP. Econ Lett 107(2):95–98

    Article  Google Scholar 

  • Stock JH, Watson MW (2002) Macroeconomic forecasting using diffusion indexes. J Econ Stat 20(2):147–162

    Article  Google Scholar 

  • Stock JH, Watson MW (2006) Forecasting with many predictors. In: Elliott G, Granger CWJ, Timmermann A (eds) Handb econ forecast, vol 1. Elsevier, Amsterdam, pp 515–554

    Chapter  Google Scholar 

  • Timmermann A (2006) Forecast combinations. In: Elliott G, Granger CWJ, Timmermann A (eds) Handb econ forecast, vol 1. Elsevier, Amsterdam, pp 135–196

    Chapter  Google Scholar 

  • Wallis KF (1986) Forecasting with an econometric model: the ‘ragged edge’ problem. J Forecast 5(1):1–13

    Article  Google Scholar 

  • Weber E, Zika G (2013) Labour market forecasting: Is disaggregation useful? IAB-discussion paper 14/2013

  • Working Group Regional Accounts VGRdL (2011) Gross domestic product, gross value added in Germany by Bundesland and East-West-Regions. Working group regional accounts VGRdL, series 1, state results volume 1, date of calculation: August 2010 / February 2011, Stuttgart 2011

Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Robert Lehmann.



Table 4 Indicators, acronyms and transformations

Rights and permissions

Reprints and permissions

About this article

Cite this article

Lehmann, R., Wohlrabe, K. Forecasting gross value-added at the regional level: are sectoral disaggregated predictions superior to direct ones?. Rev Reg Res 34, 61–90 (2014).

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


JEL Classification