Publications
Nowcasting GDP using machine learning methods
Published in AStA Advances in Statistical Analysis, 2024
This paper compares the ability of several econometric and machine learning methods to nowcast GDP in (pseudo) real-time. The analysis takes the example of Dutch GDP over the period 1992Q1–2018Q4 using a broad data set of monthly indicators. It discusses the forecast accuracy but also analyzes the use of information from the large data set of macroeconomic and financial predictors. We find that, on average, the random forest provides the most accurate forecast and nowcasts, whilst the dynamic factor model provides the most accurate backcasts. Read more
Capital and labor misallocation in the Netherlands
Published in Journal of Productivity Analysis, 2022
Using firm-level panel data we analyze the misallocation of capital and labor for the Netherlands in the period 2001-2017. We use the dispersion in marginal revenue products of capital and labor to measure the extent of misallocation. Compared to a counterfactual efficient allocation we find that misallocation has had a sizable negative impact on aggregate productivity of around 14 percentage points in the period 2001-2017. Especially capital misallocation has increased over time. Exploiting a panel data error components model we find that capital misallocation has a much more permanent character than labor misallocation. Moreover, it is the permanent component of capital misallocation that has increased over time. Finally, we show that in our sample the measurement of misallocation is largely insensitive to capital adjustment costs and alternative specifications of the production function. The contribution of heterogeneous markups to observed misallocation, however, is non-negligible. Read more
Joint decomposition of business and financial cycles: evidence from eight advanced economies
Published in Oxford Bulletin of Economics and Statistics, 2022
We discuss a model-based simultaneous decomposition of multiple time series in short-term and medium-term cyclical dynamics. We associate short-term dynamic features with the business cycle and medium-term dynamic features with the financial or credit cycle. For eight advanced economies, we analyze a set of macroeconomic and financial time series data. A strong and common finding among all economies is the co-cyclicality of medium-term cycles, especially those corresponding to house price and gross domestic product variables. We also find empirical evidence that the house price is partly driven by the financial cycle. Most cyclical movements in the country-specific time series appear to be driven by domestic rather than global factors. Read more
Combining model-based near-term GDP forecasts and judgmental forecasts: a real-time exercise for the G7 countries
Published in Oxford Bulletin of Economics and Statistics, 2018
We investigate the effects of combining model‐based near‐term GDP forecasts and judgmental (quarterly) forecasts by professional analysts (Consensus forecasts) in a real‐time setting for the G7 countries over the years 1999–2013. Model‐based forecasts are produced by a dynamic factor model (DFM). We consider as combination schemes the weighted average and the linear combination. Combining with subjective information delivers sizable gains in forecasting ability of statistical models for all countries except Japan, even when subjective forecasts are somewhat dated. Accuracy gains are much more pronounced in the volatile period after 2008 due to a marked improvement in predictive power of Consensus forecasts relative to the DFM. A possible explanation is that mechanical models may be more vulnerable to extreme observations in estimation samples. Consensus forecasts are superior at the moment of publication for most countries since 2008. For some countries forecast combinations can improve upon Consensus forecasts in between the quarterly release dates of the Consensus survey. Read more
Forecasting and nowcasting economic growth in the euro area using factor models
Published in International Journal of Forecasting, 2016
Many empirical studies provide evidence that factor models, which use a large data set of economic variables, can contribute to the computation of more accurate forecasts. In this study we examine the performance of four different factor models in a pseudo real-time forecasting competition for the euro area and five of its largest countries. Our aim is to empirically identify the best factor model approach for the forecasting and nowcasting of the quarterly Gross Domestic Product growth rate. We also propose some modifications of existing factor model specifications with the aim of empirically improving their forecast performances. We conclude that factor models consistently outperform the benchmark autoregressive model, before and during the crisis. Moreover, we find that the highest forecast accuracy is mostly produced by the collapsed dynamic factor model. Read more
Forecasting and nowcasting real GDP: comparing statistical models and subjective forecasts
Published in International Journal of Forecasting, 2016
We conduct a systematic comparison of the short-term forecasting abilities of twelve statistical models and professional analysts in a pseudo-real-time setting, using a large set of monthly indicators. Our analysis covers the euro area and its five largest countries over the years 1996–2011. We find summarizing the available monthly information in a few factors to be a more promising forecasting strategy than averaging a large number of single-indicator-based forecasts. Moreover, it is important to make use of all available monthly observations. The dynamic factor model is the best model overall, particularly for nowcasting and backcasting, due to its ability to incorporate more information (factors). Judgmental forecasts by professional analysts often embody valuable information that could be used to enhance the forecasts derived from purely mechanical procedures. Read more
Other publications
Overvest, B., Winter, J.M., Verstegen, L., Kattenberg, M. and J. Rusch (2024) Macromodellen ook waardevol in crisistijd? (Macromodels also useful in times of crisis), Economisch Statistische Berichten 109(4837), 399-401. Download here from the publisher’s website.
Pick, A. and J.M. de Winter (2023) Can machine learning methods help nowcast GDP?, SUERF Policy Brief, 521. Download here from the publisher’s website or here.
Bun, M.J.G and J.M. de Winter (2022) Misallocatie kapitaal en arbeid tijdens coronacrisis verder toegenomen (Misallocation of capital and labor increased further during the corona crisis), Economisch Statistische Berichten 107(4811), 2-5. Download here from the publisher’s website or here.
Winter, J.M. de and B. Pruijt (2022) De invloed van het corona steun- en herstel pakket op het Nederlandse bedrijfsleven (The influence of the corona support and recovery package on Dutch business), DNB Analyse, Download here.
Winter, J.M. de and D.W van Dijk (2021), Jaaroverzicht FD-sentimentsindicator: 2021 in de achteruitkijkspiegel (FD Sentiment Indicator Annual Review: 2021 in the rear-view mirror), DNB Analyse, Download here.
Winter, J.M. de and M.D. Volkerink (2021), De financiële positie van het Nederlandse mkb één jaar na de Covid-19 uitbraak (The financial position of Dutch SMEs one year after the Covid-19 outbreak), DNB Analyse , 1(2). Download here. Download all figures here.
Winter, J.M., de and D.W. van Dijk (2021), Sentimentsindicator op basis van financieel economisch nieuws (Sentimentindcator based on financial economic news), Economisch Statistische Berichten 106(4799), 340-343. Download here from the publisher’s website or here.
Bun, M.J.G., Caloia, F.G. and J.M. de Winter (2020), Hoe effectief is het COVID-19 noodpakket in het voorkomen van insolventie van bedrijven? Een stresstest van het Nederlandse MKB (How effective is the COVID-19 emergency package in preventing corporate insolvency? A stress test of Dutch SMEs), mimeo, August 11, 2020. Download here. Download appendices here.
Bun, M.J.G., and J.M. de Winter (2020), Pre-Corona geen opmars van zombies in het Nederlandse bedrijfsleven (Pre-Corona no rise of zombies in Dutch business), mimeo, March 25, 2020. Download here. Download appendices here.
Bun, M.J.G. and J.M. de Winter (2019), Misallocatie van kapitaal en arbeid in de Nederlandse economie toegenomen (Misallocation of capital and labor increased in the Netherlands), Economisch Statistische Berichten 104(4779), 514-516. Download here from the publisher’s website or here.
Winter, de J.M., M.D. Volkerink and C.H.M. Eijking (2018), Bedrijfsinvesteringen sinds crisis sterker beïnvloed door schuldpositie bedrijf (Business investment more strongly affected by firm leverage since onset of financial crisis), Economisch Statistische Berichten 103(4758), 85-87. Download here from the publisher’s website or here.
Parlevliet, J, Doll, M., Vermeulen, R. and J.M. de Winter (2016), Perspectief op groei: de Nederlandse economie in beweging (Perspectives on growth: the Dutch economy in motion), DNB Occasional Studies 14-04. Download here from the DNB-website or here.
Winter, de J.M. (2014), Kredietaanvragen geconcentreerd bij financieel zwakke bedrijven (Credit applications concentrated amongst financially weak firms), Economisch Statistische Berichten 99(4693), 566-568. Download here from the publisher’s website or here.
Veer, van der K.J.M and J.M. de Winter (2014), Herziening DNB-conjunctuurindicator (Revision of the DNB business cycle indicator), Economisch Statistische Berichten 96(4617), 525-527. Download here from the publisher’s website or here.
Hessel, J. and J.M. de Winter (2010), Verschillen in de kracht van het westerse herstel verklaard (Explaining differences in the speed of recovery in developed economies), Economisch Statistische Berichten 95(4583), 234-236. Download here from the publisher’s website or here.
Veer, van der K.J.M., and J.M. de Winter (2009), Economisch herstel eurogebied laat op zich wachten (Economic recovery euro area will take some time), Economisch Statistische Berichten 94(4562), 365-366. Download here from the publisher’s website or here.
Winter, de J.M. and K.J.M. van der Veer (2008), Voorspelkracht conjunctuurindicatoren eurogebied (Predictive power of leading indicators for the euro area), Economisch Statistische Berichten 93(4530), 136-137. Download here from the publisher;s website or here.
Winter, de J.M. and K.J.M. van der Veer (2008), Consument heeft rooskleurig beeld van de economie (Consumer has too rosy a picture of economy), Economisch Statistische Berichten 92(4515), 462-463. Download here from the publisher’s website or here.
Winter, de J.M. (2007), Loondalingen in Europa steeds normaler (Negative wage growth more common in Europe), TPEdigitaal 1(1), 46-66. Download here from the publisher’s website or here.
Winter, de J.M. (2006), Lagere inflatie omgeving in Europa: de invloed van vakbonden (Low inflation environment: the role of labor unions), Kwartaalschrift Economie 3, 274-290. Download here.
Tijdens, K.G., J.M. de Winter and J.A.C. Korteweg (2004), Voorwaarden voor verlofsparen (Requirements for leave-saving), Economisch Statistische Berichten 92(4515), 42-44. Download here from the publisher’s website or here.
Winter, de J.M. and P.H.G. Berkhout (2002), Zoekgedrag en instroom van kenniswerkers (Job search behavior and labor market entry of knowledge workers), in: Kennis en economie 2001- onderzoek en innovatie in Nederland, CBS: Voorburg/Heerlen, 41–47. Download here from the publisher’s website or here
Bijvoet, C.C, L. Bunschoten, F.A. Felsö, C.C Koopmans, J. Theeuwes and J.M de Winter (2002), De economische structuurkenmerken van de bouwnijverheid (The structural economic characteristics of the Dutch construction sector), SEO-rapportnummer 640. Download here from the website of Dutch Parliament or here.
Winter, de J.M. and W.J.J. Manshanden (2002), Statistisch onderzoek naar prijs- en kostenontwikkeling in de bouwsector (Statistical inquiry into price and cost developments in the construction sector), SEO-rapportnummer 630.
Graaf, D. de, and J.M. de Winter (2001), Effectiviteit en efficiëntie van woningcorporaties (Effectiveness and efficiency of housing corporations), SEO-rapportnummer 576.
Berkhout, E, P.H.G. Berkhout and J.M de Winter (2001), Studie en Werk 2001. HBO’ers en Academici van studiejaar 1998/1999 op de arbeidsmarkt (Education and work 2001: the labor market position of graduates in higher vocational education and university, class of 1989/1999), SEO-rapportnummer 580. Download here.
Berkhout, P.H.G., B.M.S. van Praag and J.M. de Winter, E. Edelmann and W.A. Timmerman (2000), Substitutiegedrag in de lezersmarkt van dagbladen (Substitution in the market for Dutch newspapers), SEO-rapportnummer 539.
Theeuwes, J.J.M and J.M. de Winter (1998), Econometrische evaluatie “prognose sanctiecapaciteit” (Econometric evaluation of the forecasting model for the Dutch prison system), SEO-rapportnummer 485. Download here from the website of the WODC or here.