Skip to main content
Erschienen in: Empirical Economics 2/2017

23.08.2016

Macroeconomic and credit forecasts during the Greek crisis using Bayesian VARs

verfasst von: Dimitrios P. Louzis

Erschienen in: Empirical Economics | Ausgabe 2/2017

Einloggen

Aktivieren Sie unsere intelligente Suche um passende Fachinhalte oder Patente zu finden.

search-config
loading …

Abstract

We investigate the ability of small- and medium-scale Bayesian VARs (BVARs) to produce accurate macroeconomic (output and inflation) and credit (loans and lending rate) out-of-sample forecasts during the latest Greek crisis. We implement recently proposed Bayesian shrinkage techniques based on Bayesian hierarchical modeling, and we evaluate the information content of forty-two (42) monthly macroeconomic and financial variables in terms of point and density forecasting. Alternative competing models employed in the study include Bayesian autoregressions (BARs) and time-varying parameter VARs with stochastic volatility, among others. The empirical results reveal that, overall, medium-scale BVARs enriched with economy-wide variables can considerably and consistently improve short-term inflation forecasts. The information content of financial variables, on the other hand, proves to be beneficial for the lending rate density forecasts across forecasting horizons. Both of the above-mentioned results are robust to alternative specification choices, while for the rest of the variables smaller-scale BVARs, or even univariate BARs, produce superior forecasts. Finally, we find that the popular, data-driven, shrinkage methods produce, on average, inferior forecasts compared to the theoretically grounded method considered here.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Anhänge
Nur mit Berechtigung zugänglich
Fußnoten
1
Alternatively, factor methods have also been used in studies that involve a large number of variables (e.g., see Stock and Watson, 2002, 2006 among others).
 
2
The authors also show that models with more than 20 variables can hardly improve forecasting performance.
 
3
Hyperparameter is a coefficient that is used to parameterize a prior distribution.
 
4
See Provopoulos (2014) for the origins of the Greek crisis and Gibson et al. (2014a) for an overview of the crisis in the Euro area.
 
5
Successive recession may be attributed to reduced internal demand (due to increased unemployment and reduced salaries) and weak dynamics in exports (due to firms’ reduced ability to access credit channels) (Kiriakidis and Kargas 2013).
 
6
The FSSI is a composite systemic stress index that applies the insights from the standard portfolio theory to summarize stress measures of different market segments into an aggregate index. The key feature of the FSSI is the time-varying cross-correlations among the stress measures, which form the mechanism that captures the systemic nature of stress. For an alternative version of the FSSI, see also Louzis and Vouldis (2012).
 
7
Macroeconomic variables include the industrial production index and the consumers price index, while credit variables include the outstanding amount of total loans granted to the private sector and the corresponding lending rate (see Sect. 3 and Table 7 for details).
 
8
See also the discussion in Kazanas and Tzavalis (2014).
 
9
For a recent example of the marginal approach, see Caraiani (2014).
 
10
The term conjugate prior is used to define the prior that comes from the same family of distributions along with the posterior.
 
11
An alternative approach to impose the N-IW prior is to use a fictitious prior data set as in Banbura et al. (2010).
 
12
Since the prior mean of \(\varSigma \) is defined as \(\left( {d-n-1} \right) ^{-1}\varSigma \), \(d=n+2\) is the minimum value that guarantees its existence.
 
13
The Minnesota prior was originally developed by Litterman (1979, (1980).
 
14
Hierarchical modeling refers to models that place priors on hyperparameters (i.e., hyperpriors), and thus, one more step is added to the prior structure in a hierarchical manner.
 
15
See Giannone et al. (2015) and references therein for a list of studies using these methods.
 
16
We use the industrial production index as a proxy for the aggregate output as in Caraiani (2014), since it is the only index of economic activity for Greece available at a monthly frequency. The industrial production index is also one of the main determinants of the Greek GDP (Kiriakidis and Kargas 2013).
 
17
We use the Euro OverNight Index Average (EONIA) as a proxy for the monetary policy stance in the Euro zone. The EONIA rate is defined as the average of overnight rates for unsecured interbank lending in Euro area. The Governing council of the European Central Bank (ECB) determines the range of fluctuation of the EONIA rate, which is given by the following range: [deposit facility rate, marginal lending facility rate]. The EONIA rate is considered an efficient proxy of the monetary policy stance in the Euro area, as compared to other money market instruments, such as the Euribor, the overnight interest swap (OIS) rate on EONIA rate or the repo rates (see the discussion in Ciccarelli et al. 2010; Gerlach and Lewis 2014).
 
18
As already mentioned there are also data availability issues, which pose limitations to this kind of analysis.
 
19
Estimation of the various constant parameter BVAR models is based on the MATLAB code kindly provided by Domenico Giannone in his homepage: http://​homepages.​ulb.​ac.​be/​~dgiannon/​. For the estimation of the TVP-VAR we use the MATLAB code kindly provided by Haroon Mumtaz in his homepage: https://​sites.​google.​com/​site/​hmumtaz77/​.
 
20
Recall that the baseline model uses the variables of Table 7.
 
21
In practice we set \(\lambda =10^{-4}\) and \(\lambda =10^{4}\) for the dogmatic and flat priors, respectively, as in Berg and Henzel (2015).
 
22
This is analogous to the use of principal components implemented by Banbura et al. (2010, p. 80).
 
23
The only exception is the TVP-VAR which is estimated recursively.
 
24
The description of the evaluation metrics as well as the presentation of the results follows closely the contributions of Carriero et al. (2015a, (2015b).
 
25
This is equivalent to testing whether relative RMSFE \(< 1\) and \(\Delta \) SCORE \(> 0\), respectively.
 
26
I would like to thank an anonymous reviewer for proposing to incorporate the AR model into the analysis.
 
27
Particularly, the BVAR + FSSI outranks the benchmark in 33 % of the cases, while the \(\hbox {Medium}^{\mathrm{Fin}}\)-BVAR in 17 % of the cases across variables and forecasting horizons.
 
28
The average \(\lambda \) estimates for \(\hbox {Medium}^{\mathrm{Ec}}\)-BVAR, \(\hbox {Medium}^{\mathrm{Fin}}\)-BVAR are 0.21 and 0.20 for the CKM method and 0.27 and 0.26 for the GLP method.
 
29
Other studies that implement TVP-VARs in forecasting exercises use long data sets with quarterly frequencies that cover time periods that range between 20 and 60 years (e.g., see D’Agostino et al. 2013; Barnet et al. 2014; Carriero et al. 2015a).
 
30
We use the first seventy (70) observations of our sample as an in-sample period.
 
31
The method implies rolling or recursive estimation. For \(\tau =1\) (i.e., the first point in the out-of-sample period) for which \(\lambda _{m}^{*} (\tau +1)\) is not available we set \(\lambda _{m}^{*} (1)=0.2\), which is a benchmark value used by many researchers (see Carriero et al. 2015b and references therein).
 
Literatur
Zurück zum Zitat Adolfson M, Linde J, Villani M (2007) Forecasting performance of an open economy DSGE model. Econom Rev 26:289–328CrossRef Adolfson M, Linde J, Villani M (2007) Forecasting performance of an open economy DSGE model. Econom Rev 26:289–328CrossRef
Zurück zum Zitat Amisano G, Giacomini R (2007) Comparing density forecasts via weighted likelihood ratio tests. J Bus Econ Stat 25:177–190CrossRef Amisano G, Giacomini R (2007) Comparing density forecasts via weighted likelihood ratio tests. J Bus Econ Stat 25:177–190CrossRef
Zurück zum Zitat Banbura M, Giannone D, Reichlin L (2010) Large Bayesian vector auto regressions. J Appl Econom 25:71–92CrossRef Banbura M, Giannone D, Reichlin L (2010) Large Bayesian vector auto regressions. J Appl Econom 25:71–92CrossRef
Zurück zum Zitat Bank of Greece, 2014 (2013) Stress test of the Greek banking sector. Bank of Greece March 2014 Bank of Greece, 2014 (2013) Stress test of the Greek banking sector. Bank of Greece March 2014
Zurück zum Zitat Barnet A, Mumtaz H, Theodoridis K (2014) Forecasting UK GDP growth and inflation under structural change. A comparison of models with time-varying parameters. Int J Forecast 30:129–143CrossRef Barnet A, Mumtaz H, Theodoridis K (2014) Forecasting UK GDP growth and inflation under structural change. A comparison of models with time-varying parameters. Int J Forecast 30:129–143CrossRef
Zurück zum Zitat Berg TO, Henzel SR (2015) Point and density forecasts for the euro area using Bayesian VARs. Int J Forecast 31:1067–1095CrossRef Berg TO, Henzel SR (2015) Point and density forecasts for the euro area using Bayesian VARs. Int J Forecast 31:1067–1095CrossRef
Zurück zum Zitat Blanchard O, Leigh D (2013) Growth forecast errors and fiscal multipliers. IMF working paper WP/13/1 Blanchard O, Leigh D (2013) Growth forecast errors and fiscal multipliers. IMF working paper WP/13/1
Zurück zum Zitat Caraiani P (2014) Do money and financial variables help forecasting output in emerging European economies? Empir Econ 46:743–763CrossRef Caraiani P (2014) Do money and financial variables help forecasting output in emerging European economies? Empir Econ 46:743–763CrossRef
Zurück zum Zitat Carriero A, Kapetanios G, Marcellino M (2009) Forecasting exchange rates with large Bayesian VAR. Int J Forecast 25:400–417CrossRef Carriero A, Kapetanios G, Marcellino M (2009) Forecasting exchange rates with large Bayesian VAR. Int J Forecast 25:400–417CrossRef
Zurück zum Zitat Carriero A, Kapetanios G, Marcellino M (2012) Forecasting government bond yields with large Bayesian vector autoregressions. J Bank Finance 36:2026–2047CrossRef Carriero A, Kapetanios G, Marcellino M (2012) Forecasting government bond yields with large Bayesian vector autoregressions. J Bank Finance 36:2026–2047CrossRef
Zurück zum Zitat Carriero A, Mumtaz H, Theophilopoulou A (2015a) Macroeconomic information, structural change, and the prediction of fiscal aggregates. Int J Forecast 31:325–348CrossRef Carriero A, Mumtaz H, Theophilopoulou A (2015a) Macroeconomic information, structural change, and the prediction of fiscal aggregates. Int J Forecast 31:325–348CrossRef
Zurück zum Zitat Carriero A, Clarck TE, Marcellino M (2015b) Bayesian VARs: specification choices and forecast accuracy. J Appl Econom 30:46–73CrossRef Carriero A, Clarck TE, Marcellino M (2015b) Bayesian VARs: specification choices and forecast accuracy. J Appl Econom 30:46–73CrossRef
Zurück zum Zitat Ciccarelli M, Maddaloni A, Peydro J-L (2010) Trusting the bankers: a new look at the credit channel of monetary policy. ECB working papers series no 1228, July 2010 Ciccarelli M, Maddaloni A, Peydro J-L (2010) Trusting the bankers: a new look at the credit channel of monetary policy. ECB working papers series no 1228, July 2010
Zurück zum Zitat Clark T, McCracken M (2013) Advances in forecast evaluation. In: Elliot G, Timmermann A (eds) Handbook of economic forecasting, vol 2(B). Elsevier, Amsterdam, pp 1107–1201 Clark T, McCracken M (2013) Advances in forecast evaluation. In: Elliot G, Timmermann A (eds) Handbook of economic forecasting, vol 2(B). Elsevier, Amsterdam, pp 1107–1201
Zurück zum Zitat D’Agostino A, Gambetti L, Giannone D (2013) Macroeconomic forecasting and structural change. J Appl Econom 28:82–101CrossRef D’Agostino A, Gambetti L, Giannone D (2013) Macroeconomic forecasting and structural change. J Appl Econom 28:82–101CrossRef
Zurück zum Zitat De Mol C, Giannone D, Reichlin L (2008) Forecasting using a large number of predictors: is Bayesian regression a valid alternative to principal components? J Econom 146:318–328CrossRef De Mol C, Giannone D, Reichlin L (2008) Forecasting using a large number of predictors: is Bayesian regression a valid alternative to principal components? J Econom 146:318–328CrossRef
Zurück zum Zitat Diebold FX, Mariano RS (1995) Comparing predictive accuracy. J Bus Econ Stat 13:253–263 Diebold FX, Mariano RS (1995) Comparing predictive accuracy. J Bus Econ Stat 13:253–263
Zurück zum Zitat Doan T, Litterman R, Sims C (1984) Forecasting and conditional projection using realistic prior distributions. Econom Rev 3:1–144CrossRef Doan T, Litterman R, Sims C (1984) Forecasting and conditional projection using realistic prior distributions. Econom Rev 3:1–144CrossRef
Zurück zum Zitat Faust J, Wright JH (2009) Comparing Greenbook and reduced form forecasts using a large realtime dataset. J Bus Econ Stat 27:468–479CrossRef Faust J, Wright JH (2009) Comparing Greenbook and reduced form forecasts using a large realtime dataset. J Bus Econ Stat 27:468–479CrossRef
Zurück zum Zitat Geweke J, Amisano G (2010) Comparing and evaluating Bayesian predictive distributions of asset returns. Int J Forecast 26:216–230CrossRef Geweke J, Amisano G (2010) Comparing and evaluating Bayesian predictive distributions of asset returns. Int J Forecast 26:216–230CrossRef
Zurück zum Zitat Giannone D, Lenza M, Momferatou D (2014) Short-term inflation projections: a Bayesian vector autoregressive approach. Int J Forecast 30:635–644CrossRef Giannone D, Lenza M, Momferatou D (2014) Short-term inflation projections: a Bayesian vector autoregressive approach. Int J Forecast 30:635–644CrossRef
Zurück zum Zitat Giannone D, Lenza M, Primiceri G (2015) Prior selection for vector autoregressions. Rev Econ Stat 97:436–451CrossRef Giannone D, Lenza M, Primiceri G (2015) Prior selection for vector autoregressions. Rev Econ Stat 97:436–451CrossRef
Zurück zum Zitat Gibson HD, Hall SG, Tavlas GS (2012) The Greek financial crisis: growing imbalances and sovereign spreads. J Int Money Finance 31(3):498–516CrossRef Gibson HD, Hall SG, Tavlas GS (2012) The Greek financial crisis: growing imbalances and sovereign spreads. J Int Money Finance 31(3):498–516CrossRef
Zurück zum Zitat Gibson HD, Palivos T, Tavlas GS (2014a) The crisis in the Euro area: an analytic overview. J Macroecon 39:233–239CrossRef Gibson HD, Palivos T, Tavlas GS (2014a) The crisis in the Euro area: an analytic overview. J Macroecon 39:233–239CrossRef
Zurück zum Zitat Gibson HD, Hall SG, Tavlas GS (2014b) Fundamentally wrong: market pricing of sovereigns and the Greek financial crisis. J Macroecon 39:405–419CrossRef Gibson HD, Hall SG, Tavlas GS (2014b) Fundamentally wrong: market pricing of sovereigns and the Greek financial crisis. J Macroecon 39:405–419CrossRef
Zurück zum Zitat Gerlach S, Lewis J (2014) Zero lower bound, ECB interest rate policy and the financial crisis. Empir Econ 46(3):865–886CrossRef Gerlach S, Lewis J (2014) Zero lower bound, ECB interest rate policy and the financial crisis. Empir Econ 46(3):865–886CrossRef
Zurück zum Zitat Gupta R, Kabundi A (2010) Forecasting macroeconomic variables in a small open economy: a comparison between small- and large-scale models. J Forecast 29:168–185CrossRef Gupta R, Kabundi A (2010) Forecasting macroeconomic variables in a small open economy: a comparison between small- and large-scale models. J Forecast 29:168–185CrossRef
Zurück zum Zitat Harvey D, Leybourne S, Newbold P (1997) Testing the equality of prediction mean squared errors. Int J Forecast 13:281–291CrossRef Harvey D, Leybourne S, Newbold P (1997) Testing the equality of prediction mean squared errors. Int J Forecast 13:281–291CrossRef
Zurück zum Zitat Hristov N, Hulsewig O, Wollmershauser T (2012) Loan supply shocks during the financial crisis: evidence for the Euro area. J Int Money Finance 31:569–592CrossRef Hristov N, Hulsewig O, Wollmershauser T (2012) Loan supply shocks during the financial crisis: evidence for the Euro area. J Int Money Finance 31:569–592CrossRef
Zurück zum Zitat Kadiyala K, Karlsson S (1997) Numerical methods for estimation and inference in Bayesian VAR models. J Appl Econom 12:99–132CrossRef Kadiyala K, Karlsson S (1997) Numerical methods for estimation and inference in Bayesian VAR models. J Appl Econom 12:99–132CrossRef
Zurück zum Zitat Karlsson S (2013) Forecasting with Bayesian vector autoregression. Handb Econ Forecast 2:791–897CrossRef Karlsson S (2013) Forecasting with Bayesian vector autoregression. Handb Econ Forecast 2:791–897CrossRef
Zurück zum Zitat Kazanas T, Tzavalis E (2014) Comment on “Fundamentally wrong: market mispricing of sovereigns and the Greek financial crisis”. J Macroecon 39:420–423CrossRef Kazanas T, Tzavalis E (2014) Comment on “Fundamentally wrong: market mispricing of sovereigns and the Greek financial crisis”. J Macroecon 39:420–423CrossRef
Zurück zum Zitat Kiriakidis M, Kargas A (2013) Greek GDP forecast estimates. Appl Econ Lett 20:767–772CrossRef Kiriakidis M, Kargas A (2013) Greek GDP forecast estimates. Appl Econ Lett 20:767–772CrossRef
Zurück zum Zitat Koop G, Korobilis D (2010) Bayesian multivariate time series methods for empirical macroeconomics. Foun Trends Econom 3:267–358CrossRef Koop G, Korobilis D (2010) Bayesian multivariate time series methods for empirical macroeconomics. Foun Trends Econom 3:267–358CrossRef
Zurück zum Zitat Koop G (2013) Forecasting with medium and large Bayesian VARs. J Appl Econom 28:177–203CrossRef Koop G (2013) Forecasting with medium and large Bayesian VARs. J Appl Econom 28:177–203CrossRef
Zurück zum Zitat Litterman RB (1979) Techniques of forecasting using vector autoregressions. Federal Reserve of Minneapolis working paper 115 Litterman RB (1979) Techniques of forecasting using vector autoregressions. Federal Reserve of Minneapolis working paper 115
Zurück zum Zitat Litterman RB (1980) A Bayesian procedure for forecasting with vector autoregressions. Working paper, Massachusetts Institute of Technology, Department of Economics Litterman RB (1980) A Bayesian procedure for forecasting with vector autoregressions. Working paper, Massachusetts Institute of Technology, Department of Economics
Zurück zum Zitat Litterman RB (1986) Forecasting with Bayesian vector autoregressions: five years of experience. J Bus Econ Stat 4:25–38 Litterman RB (1986) Forecasting with Bayesian vector autoregressions: five years of experience. J Bus Econ Stat 4:25–38
Zurück zum Zitat Louzis DP, Vouldis AT (2012) A methodology for constructing a financial systemic stress index: an application to Greece. Econ Model 29:1228–1241CrossRef Louzis DP, Vouldis AT (2012) A methodology for constructing a financial systemic stress index: an application to Greece. Econ Model 29:1228–1241CrossRef
Zurück zum Zitat Louzis DP, Vouldis AT (2013) A financial systemic stress index for Greece. ECB working paper series no 1563, July 2013 Louzis DP, Vouldis AT (2013) A financial systemic stress index for Greece. ECB working paper series no 1563, July 2013
Zurück zum Zitat Louzis DP (2015) Measuring spillover effects in Euro area financial markets: a disaggregate approach. Empir Econ 49:1367–1400CrossRef Louzis DP (2015) Measuring spillover effects in Euro area financial markets: a disaggregate approach. Empir Econ 49:1367–1400CrossRef
Zurück zum Zitat Marcellino M (2008) A linear benchmark for forecasting GDP growth and inflation? J Forecast 27:305–340CrossRef Marcellino M (2008) A linear benchmark for forecasting GDP growth and inflation? J Forecast 27:305–340CrossRef
Zurück zum Zitat Provopoulos G (2014) The Greek economy and banking system: recent developments and the way forward. J Macroecon 39:240–249CrossRef Provopoulos G (2014) The Greek economy and banking system: recent developments and the way forward. J Macroecon 39:240–249CrossRef
Zurück zum Zitat Sims C (1993) A nine-variable probabilistic macroeconomic forecasting model. In: Business cycles, indicators and forecasting, nber chapters. National Bureau of Economic Research, Inc. pp 179–212 Sims C (1993) A nine-variable probabilistic macroeconomic forecasting model. In: Business cycles, indicators and forecasting, nber chapters. National Bureau of Economic Research, Inc. pp 179–212
Metadaten
Titel
Macroeconomic and credit forecasts during the Greek crisis using Bayesian VARs
verfasst von
Dimitrios P. Louzis
Publikationsdatum
23.08.2016
Verlag
Springer Berlin Heidelberg
Erschienen in
Empirical Economics / Ausgabe 2/2017
Print ISSN: 0377-7332
Elektronische ISSN: 1435-8921
DOI
https://doi.org/10.1007/s00181-016-1128-y

Weitere Artikel der Ausgabe 2/2017

Empirical Economics 2/2017 Zur Ausgabe