A method for taking models to the data

Department of Economics, Boston College, 140 Commonwealth Avenue, Chestnut Hill, MA 02467-3806, USA
Journal of Economic Dynamics and Control (Impact Factor: 0.86). 02/2004; 28(6):1205-1226. DOI: 10.1016/S0165-1889(03)00080-0
Source: RePEc

ABSTRACT This paper develops a method for combining the power of a dynamic, stochastic, general equilibrium model with the flexibility of a vector autoregressive time-series model to obtain a hybrid that can be taken directly to the data. It estimates this hybrid model via maximum likelihood and uses the results to address a number of issues concerning the ability of a prototypical real business cycle model to explain movements in aggregate output and employment in the postwar US economy, the stability of the real business cycle model's structural parameters, and the performance of the hybrid model's out-of-sample forecasts.

  • Source
    • "These models appear to be particularly suited for conducting policy evaluation, as shown in the works of Smets and Wouters (2003, 2004), Del Negro and Schorfheide (2004), Adolfson et al. (2008) and Christiano et al. (2005). However, calibrated DSGE models face many important challenges such as the fragility of parameter estimates, statistical fit and the weak reliability of policy forecasts as reported in Stock and Watson (2001), Ireland (2004) and Schorfheide (2010). In recent years Bayesian estimation has become popular mainly because it provides a system-based estimation approach that offers the advantage of employing prior assumptions about the parameters based on economic theory. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Advanced Bayesian methods are employed in estimating dynamic stochastic general equilibrium (DSGE) models. Although policymakers and practitioners are particularly interested in DSGE models, these are typically too stylized to be taken directly to the data and often yield weak prediction results. Hybrid models can deal with some of the DSGE model misspecifications. Major advances in Bayesian estimation methodology could allow these models to outperform well-known time series models and effectively deal with more complex real-world problems as richer sources of data become available. A comparative evaluation of the out-of-sample predictive performance of many different specifications of estimated DSGE models and various classes of VAR models is performed, using datasets from the US economy. Simple and hybrid DSGE models are implemented, such as DSGE–VAR and Factor Augmented DSGEs and tested against standard, Bayesian and Factor Augmented VARs. Moreover, small scale models including the real gross domestic product, the harmonized consumer price index and the nominal short-term federal funds interest rate, are comparatively assessed against medium scale models featuring additionally sticky nominal prices, wage contracts, habit formation, variable capital utilization and investment adjustment costs. The investigated period spans 1960:Q4–2010:Q4 and forecasts are produced for the out-of-sample testing period 1997:Q1–2010:Q4. This comparative validation can be useful to monetary policy analysis and macro-forecasting with the use of advanced Bayesian methods.
    Computational Statistics & Data Analysis 03/2014; 71:298–323. DOI:10.1016/j.csda.2013.09.018 · 1.15 Impact Factor
  • Source
    • "In particular, we fix three parameters β, δ and B. As explained in Altig et al. (2011) and Ireland (2004), it is necessary to calibrate the discount factor to successfully estimate the remaining parameters of the real business cycle model. This is particularly relevant in this setup with no capital accumulation. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Recent empirical evidence establishes that a positive technology shock leads to a decline in labor inputs. Standard RBC models fails to replicate this stylized fact, while recent papers show that augmenting the model with implementation lags, or habit formation, or shock persistence in growth rates among others accounts for this fact. In this paper, we show that a standard flexible price model with labor market frictions that allows hiring costs to depend on technology shocks may also lead to the same negative impact on labor inputs. Labor market frictions are therefore able to account for the fall in labor inputs. However, the elasticity of hiring costs to technology shocks is large, suggesting that additional extensions to the model are needed.
    Labour Economics 01/2013; 26. DOI:10.1016/j.labeco.2013.11.004 · 0.92 Impact Factor
  • Source
    • "DSGE models have been considered as forecasting tools only since the seminal study of Smets and Wouters (2003, 2004). Calibrated DSGE models often yield fragile results, when traditional econometric methods are used for estimation (Smets and Wouters 2003; Ireland 2004). Following this idea of combining the DSGE model information and the VAR representation, among other models that have been proposed in the literature, in this study we use the DSGE–VAR hybrid model. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Over the last few years, there has been a growing interest in DSGE modelling for predicting macroeconomic fluctuations and conducting quantitative policy analysis. Hybrid DSGE models have become popular for dealing with some of the DSGE misspecifications as they are able to solve the trade-off between theoretical coherence and empirical fit. However, these models are still linear and they do not consider time variation for parameters. The time-varying properties in VAR or DSGE models capture the inherent nonlinearities and the adaptive underlying structure of the economy in a robust manner. In this article, we present a state-space time-varying parameter VAR model. Moreover, we focus on the DSGE–VAR that combines a microfounded DSGE model with the flexibility of a VAR framework. All the aforementioned models as well simple DSGEs and Bayesian VARs are used in a comparative investigation of their out-of-sample predictive performance regarding the US economy. The results indicate that while in general the classical VAR and BVARs provide with good forecasting results, in many cases the TVP–VAR and the DSGE–VAR outperform the other models.
    Empirical Economics 01/2013; 45(1). DOI:10.1007/s00181-012-0623-z · 0.60 Impact Factor
Show more


Available from