Ard den Reijer

A criterion for the number of factors in a data-rich environment, 2015 (with J. Jacobs and P. Otter). (abstract) (pdf)

Coordination versus flexibility in wage formation: a focus on the nominal wage impact of productivity in Germany, Greece, Ireland, Portugal, Spain and the United States, Applied Economics 46(7), 2014 (with M. Peeters). (abstract) (pdf) (doi)

Forecasting Dutch GDP and inflation using alternative factor model specifications based on large and small data sets, Empirical Economics 44(2), 2013. (abstract) (pdf) (ppt) (doi)

MOSES: Model for Studying the Economy of Sweden, Economic Modelling 29(6), 2012 (with G. Bårdsen, P. Jonasson and R. Nymoen). (abstract) (pdf) (ppt) (doi)

Information, data dimension, and factor structure, Journal of Multivariate Analysis 106, 2012. (with J.P.A.M. Jacobs and P.W. Otter). (abstract) (pdf) (ppt) (doi)

Identifying Regional and Sectoral Dynamics of the Dutch Staffing Labour Cycle, Economic Modelling 28(4), 2011. (abstract) (pdf) (ppt) (doi)

The Dutch business cycle: a finite sample approximation of selected leading indicators, Journal of Business Cycle Measurement and Analysis 2009(2), 2009. (abstract) (pdf) (ppt) (doi)

Short-term forecasting of GDP using large monthly datasets: a pseudo real-time forecast evaluation exercise, Journal of Forecasting 28(7), 2009 (with G. Rünstler, K. Barhoumi, S. Benk, R. Cristadoro, A. Jakaitiene, P. Jelonek, A. Rua, K. Ruth and C. Van Nieuwenhuyze). (abstract) (pdf) (doi)

On wage formation, wage development and flexibility: a comparison between european countries and the United States, Applied Econometrics and International Developments 8(1), 2008 (with H. Peeters). (abstract) (pdf) (ppt) (ssrn)

Deviation cycles in manufacturing: business cycle measurement and leading indicators, Journal of Business Cycle Measurement and Analysis 2007(1), 2007. (abstract) (pdf) (ppt) (doi)

Dutch GDP data revisions: are they predictable and where do they come from? Applied Economics Quarterly 52(4), 2006 (with O. Roodenburg). (abstract) (pdf) (ppt) (ssrn)

Forecasting inflation: An art as well as a science!, De Economist 154(1), 2006 (with P. Vlaar). (abstract) (pdf) (ppt) (doi)

Nowcasting, Penning- och Valutapolitik, no. 1, 2015, p. 73-86. (English pdf)(Svenska pdf)

The Dutch Business Cycle: Which Indicators Should We Monitor?, AENorm: Magazine for students in Actuarial Sciences, Econometrics & Operations Research, no. 62, 2009, p. 74-78. (pdf)

Book Review: Modeling Aggregate Behaviour and Fluctuations in Economics: Stochastic Views of Interacting Agents, by M. Aoki, De Economist 150(5), 2002, pp. 622-623. (pdf)

De Nederlandsche Bank volgt conjunctuur gebrekkig, Me Judice, 17 oktober 2011. (pdf)

Promotierubriek: Macro-economisch voorspellen op basis van voorlopende conjunctuurindicatoren, Economisch Statistische Berichten, nr. 4588, 2010. (pdf)

Schommelingen in het consumentenvertrouwen, Economisch Statistische Berichten, nr. 4516, 2007. (pdf)

Industriële productie nader bezien, Economisch Statistische Berichten, nr. 4506, 2007. (pdf)

Nederlandse hoogconjunctuur nader beschouwd, Economisch Statistische Berichten, nr. 4500, 2006. (pdf)

Betere vooruitzichten ondanks vlakkere yieldcurve, Economisch Statistische Berichten, nr. 4482, 2006. (pdf)

De uitvoerconjunctuur, Economisch Statistische Berichten, nr. 4480, 2006. (pdf)

De consumptieconjunctuur, Economisch Statistische Berichten nr. 4478, 2006. (pdf)

De uitzendconjunctuur, Economisch Statistische Berichten nr. 4470, 2005. (pdf)

Business cycles as usual?, Economisch Statistische Berichten nr. 4460, 2005. (pdf)

Het koopgedrag van de consument, Economisch Statistische Berichten nr. 4451, 2005. (pdf)

De conjunctuur, de indicator en de waarheid, Economisch Statistische Berichten nr. 4441, 2004. (pdf)

Jetzt geht´s hausse?, Economisch Statistische Berichten nr. 4437, 2004. (pdf)

Conjunctuurindicatoren bij DNB, Economisch Statistische Berichten nr. 4425, 2004.(pdf)

De Nederlandse conjunctuur in spagaat, Economisch Statistische Berichten nr. 4409, 2003.(pdf)

De wereldconjunctuur bestaat, Economisch Statistische Berichten, nr. 4401, 2003.(pdf)

Verschillende economieën, dezelfde conjunctuurbeweging?, DNB Kwartaalbericht DNB, 2002. (pdf)

A criterion for the number of factors in a data-rich environment, 2015 (with J. Jacobs and P. Otter). (abstract) (pdf)

Abstract: This paper derives a new criterion for the determination of the number of factors in static approximate factor models, that is strongly associated with the scree test. Our criterion looks for the number of eigenvalues for which the difference between adjacent eigenvalue - eigenvalue component number blocks is maximized. Monte Carlo experiments compare the properties of our criterion to the Edge Distribution (ED) estimator of Onatski (2010) and the two eigenvalue ratio estimators of Ahn and Horenstein (2013). Our criterion outperforms the latter two for all sample sizes and the ED estimator of Onatski (2010) for samples up to 300 variables/observations.

Coordination versus flexibility in wage formation: a focus on the nominal wage impact of productivity in Germany, Greece, Ireland, Portugal, Spain and the United States, Applied Economics (forthcoming) (with M. Peeters). (abstract) (pdf) (ssrn)

Abstract: Wage coordination between countries of the European Monetary Union (EMU) aims at aligning nominal wage growth with labour productivity growth at the national level. We analyse the developments in Germany, the EMU's periphery countries Greece, Ireland, Portugal and Spain along with the US over the period 1980-2010. Apart from the contribution of productivity to nominal wages, we take into account the contributions of prices, unemployment, replacement rates and taxes by means of an econometrically estimated non-linear wage equation resulting from a wage bargaining model. We further study the downward rigidities of nominal wages. The findings show that in past times of low productivity, price inflation and reductions in unemployment put significant upward pressure on wage growth. The periphery countries are far from aligning nominal wage growth with productivity growth. German productivity is a major wage determinant, but surely not the only one. Within the context of a free bargaining between employers and labour unions, policy makers can effectively use the replacement rate to steer the nominal wage outcome.

MOSES: Model for Studying the Economy of Sweden, Economic Modelling (forthcoming) (with G. Bårdsen, P. Jonasson and R. Nymoen). (abstract) (pdf) (ppt) (doi)

Abstract: MOSES is an aggregate econometric model for Sweden, estimated on quarterly data, and intended for short-term forecasting and policy simulations. After a presentation of qualitative model properties, the econometric methodology is summarized. The model properties, within sample simulations, and examples of dynamic simulation (model forecasts) for the period 2009q2-2012q4 are presented. We address practical issues relating to operational use and maintenance of a macro model of this type. The detailed econometric equations are reported in an appendix.

Presentation at a Riksbank seminar can be found here.

Forecasting Dutch GDP and inflation using alternative factor model specifications based on large and small data sets, Empirical Economics (forthcoming). (abstract) (pdf) (ppt) (doi)

Abstract: We compare the factor forecasting performance of nested specifications of the generalized factor model based on various configurations of a large macroeconomic data set. The forecast simulation design involves in-sample model selection, factor estimation, parameter estimation and, finally, generating factor forecasts and factor augmented autoregressive forecasts. In order to empirically determine the importance of the size and the structure of the data set, we run the forecast simulation design for different configurations of the data set. We compare the factor model diagnostics of each specification and data configuration with the corresponding forecast performance. The results favour the factor structure as the specification that imposes the factor structure to the least extent and, hence, is allowed most flexibility to adapt to the data, is significantly being outperformed. Moreover, the results show that size matters as though smaller macroeconomic data sets exhibit stronger coherence, the factors being well fit do, however, generally not show improved forecasting performance.

Presentation of this paper given at a DNB lunch seminar, at Sveriges Riksbank, Deutsche Bundesbank, Tilburg University, University of Groningen and the 20th Annual Conference of the European Economic Association (EEA) in Amsterdam 2005 can be found here.

Information, data dimension, and factor structure, Journal of Multivariate Analysis 106, 2012 (with J.P.A.M. Jacobs and P.W. Otter). (abstract) (pdf) (ppt) (doi)

Abstract: This paper employs concepts from information to choosing the dimension of a data set. We propose a relative information measure connected to Kullback-Leibler numbers. By ordering the series of the data according to the measure, we are able to obtain a subset of a data set that is most informative. The method can be used as a first step in the construction of a dynamic factor model or a leading index, as illustrated with a Monte Carlo study and with the U.S. data set of Stock & Watson (1998).

Presentation of this paper at the 13th International Conference on Computing in Economics and Finance, Montréal 2007 and my presentation at the Centre for Central Banking Studies of the Bank of England, at the conference on Factor Structures for Panel and Multivariate Time Series Data at Maastricht University and at an DNB lunch seminar can be found here.

Identifying Regional and Sectoral Dynamics of the Dutch Staffing Labour Cycle, Economic Modelling 28(4), 2011. (abstract) (pdf) (ppt) (doi)

Abstract: This study analyses the dynamic characteristics of staffing employment across different business sectors and across different geographical regions in the Netherlands. We analyse a micro data set of the market leader of the Dutch staffing employment market, i.e. Randstad. We apply the dynamic factor model to extract common information out of a large data set and to isolate business cycle frequencies with the aim of forecasting economic activity. We identify regions and sectors whose cyclical developments lead the staffing labour cycle at the country level. The second question is then which model specification can exploit best the identified leading indicators at the disaggregate level to forecast the country aggregate? The dynamic factor model turns out to outperform univariate benchmark forecasting models by exploiting the substantial temporal variation of the staffing labour market at the disaggregate level.

Presentation of this paper at seminars at De Nederlandsche Bank and the 5th studiedag conjuntuur at Nyenrode University can be found here.

The Dutch business cycle: a finite sample approximation of selected leading indicators, Journal of Business Cycle Measurement and Analysis 2009(2), 2009. (abstract) (pdf) (ppt) (doi)

Abstract: In this study we construct a business cycle indicator for the Netherlands. The Christiano-Fitzgerald bandpass filter is employed to isolate the cycle using the definition of business cycle frequencies as waves with lengths longer than 3 years and shorter than 11 years. The coincident business cycle index is based on industrial production, household consumption and staffing employment. These three variables represent key macroeconomic developments, which are also analysed by both the CEPR and NBER dating committees. The composite leading index consists of eleven indicators representing different sectors in the economy: three financial series, four business and consumer surveys and four real activity variables, of which two supply- and two demand-related. The pseudo real-time performance of the composite indicator is analyzed by the extent to which the indicator gets revised as more data becomes available. Finally, the composite leading indicator is employed in a bivariate Vector Autoregressive model to forecast GDP growth rates.

Presentation of this paper given at a DNB lunch seminar, the 4th Studiedag Conjunctuur, the 28th CIRET Conference on Cyclical Indicators and Economic Policy Decisions in Rome, the International Workshop on Computational and Financial Econometrics (CFE 2007) in Geneva, the 27th Annual International Symposium on Forecasting on Financial Forecasting in a Global Economy in New York City and a seminar at The Conference Board in New York City can be found here. The publication of the indicator in the Dutch financial newspaper Het Financieele Dagblad on the 25th of September 2007 can be found here. The indicator is presented in a not-too-technical way in a magazine for econometrics students and can be found here. Finally, the monthly update of the indicator has been published on the website of De Nederlandsche Bank and on the website of the Dutch policy journal Economisch Statistische Berichten from mid 2006 until August 2011. During this period, the business cycle indicator is mistakenly critized for late signalling of turning points as explained here (in Dutch) (pdf).

Short-term forecasting of GDP using large monthly datasets: a pseudo real-time forecast evaluation exercise, Journal of Forecasting 28(7), 2009 (with G. Rünstler, K. Barhoumi, S. Benk, R. Cristadoro, A. Jakaitiene, P. Jelonek, A. Rua, K. Ruth and C. Van Nieuwenhuyze). (abstract) (pdf) (doi)

Abstract: This paper evaluates different models for the short-term forecasting of real GDP growth in ten selected European countries and the euro area as a whole. Purely quarterly models are compared with models designed to exploit early releases of monthly indicators for the nowcast and forecast of quarterly GDP growth. Amongst the latter, we consider small bridge equations and forecast equations in which the bridging between monthly and quarterly data is achieved through a regression on factors extracted from large monthly datasets. The forecasting exercise is performed in a simulated real-time context, which takes account of publication lags in the individual series. In general, we fi nd that models that exploit monthly information outperform models that use purely quarterly data and, amongst the former, factor models perform best..

On wage formation, wage development and flexibility: a comparison between european countries and the United States, Applied Econometrics and International Developments 8(1), 2008 (with H. Peeters). (abstract) (pdf) (ppt) (ssrn)

Abstract: For Germany, Spain, France, the Netherlands and the US an Error Correction Model with a long-term non-linear wage equation is estimated by 3-SLS to obtain consistent estimates, accounting for endogeneity and common shocks. On the basis of the estimated parameter elasticities of wages with respect to labour productivity, value added and consumer prices, taxes, unemployment and replacement rates are computed along with the wage contributions. The results indicate that the dominant role of prices in the formation of wages in the seventies and eighties was taken over by labour productivity in the US and unemployment in Spain and – almost- in the Netherlands at the end of the nineties. Evidence for a stronger real wage flexibility of the US in comparison with the four European countries is not found.

Presentation of this paper at seminars at De Nederlandsche Bank, European Central Bank, the European Commission Directorate General Employment and Social Affairs, Banque de France Symposium on Modelling and Forecasting wage and price dynamics in France, the US and the euro area, a poster session at the 18th Annual Conference of the European Economic Association (EEA) in Stockholm 2003 can be found here.

Deviation cycles in manufacturing: business cycle measurement and leading indicators, Journal of Business Cycle Measurement and Analysis 2007(1), 2007. (abstract) (pdf) (ppt) (doi)

Abstract:The deviation cycles in the manufacturing industry of nine OECD-countries are identified by applying the Christiano-Fitzgerald bandpass filter. Turning points, low- and high-growth phases and other descriptive statistics are derived from these deviation cycles. A regression based test statistic is applied to test for duration dependence. Moreover, the international linkage between the cyclical motions in the manufacturing industry of two countries is investigated by measuring the degree of synchronisation. In addition to measuring the cyclical fluctuation, a composite leading indicator is constructed which replicates and predicts the deviation cycle in the manufacturing industry. This composite leading indicator is a single index composed of economic, financial and expectation variables possessing leading properties.

Presentation of this paper given at a DNB lunch seminar, the Econometric Society European Meeting (ESEM) 2003 in Stockholm, SOM Workshop on business cycle analysis at the university of Groningen and a seminar at Sveriges Riksbanken can be found here.

Dutch GDP data revisions: are they predictable and where do they come from? Applied Economics Quarterly 52(4), 2006 (with O. Roodenburg). (abstract) (pdf) (ppt) (ssrn)

Abstract:This paper examines whether the preliminary releases of GDP incorporate efficiently all available information or whether the preliminary estimates contain information that can be useful in predicting forthcoming GDP data revisions. Forecast rationality tests are applied to distinguish between these two characterisations. We analyse the revision over three horizons: the very short term revision after one quarter, the short-term revision after two years and the long-term revision. We find evidence of predictability for all short- and long-term revisions of Dutch GDP data. Our evidence for the revisions of the seasonally adjusted quarter-on-quarter growth rates are in line with the findings of G7 countries. Moreover, we analyse the revisions of the six expenditure components and ten prodcution components that constitute GDP. Only the preliminary releases of household consumption and the construction sector seem to explain the GDP data revisions. However, the general conclusion is that the forecast rationality hypothesis is rejected for almost all components separately, while almost no individual component's preliminary data release can forecast the revision of GDP.

Presentation of this paper given at a DNB lunch seminar and at the 5th EABCN/CEPR Workshop Needed: A Real-Time Database for the Euro Area can be found here.

Forecasting inflation: An art as well as a science!, De Economist 154(1), 2006 (with P. Vlaar). (abstract) (pdf) (ppt) (doi)

Abstract: In this study, we build two forecasting models to predict inflation Harmonised Index of Consumer Prices (HICP) for the Netherlands and for the euro area. The models provide point forecasts and prediction intervals for both the components of the HICP and the aggregated HICP-index itself. Both models are small-scale linear time series models allowing for long-run equilibrium relationships between HICP components and other variables, notably the hourly wage rate and the import or producer prices. The model for the Netherlands is used to generate the Dutch inflation projections for the eurosystem’s Narrow Inflation Projection Exercise (NIPE). The recursive forecast errors for several forecast horizons are evaluated for all models, and are found to outperform a naive forecast and optimal AR models. Moreover, the same result holds for the Dutch NIPE projections, which have been provided quarterly since 1999. The aggregation method to predict total HICP inflation generally outperforms the direct method, except for long horizons in the case of the Netherlands.

Presentation of this paper given at the 1st EABCN/CEPR Conference on the Euro Area Business Cycle, the 10th International Conference on Computing in Economics and Finance in Amsterdam, a DNB lunch seminar and at Erasmus University Rotterdam and Maastricht University can be found here.