On the predictability of time-varying VAR and DSGE models

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.     

Forecasting the semiconductor industry cycles by bootstrap prediction intervals

In recent years, there has been a recognition that point forecasts of the semiconductor industry sales may often be of limited value. There is substantial interest for a policy maker or an individual investor in knowing the degree of uncertainty that attaches to the point forecast before deciding whether to increase production of semiconductors or purchase a particular share from the semiconductor stock market. In this article, I first obtain the bootstrap prediction intervals of the global semiconductor industry cycles by a vector autoregressive (VAR) model using monthly US data consisting of four macroeconomic and seven industry-level variables with 92 observations. The 24-step-ahead out-of-sample forecasts from various VAR setups are used for comparison. The empirical result shows that the proposed 11-variable VAR model with the appropriate lag length captures the cyclical behaviour of the industry and outperforms other VAR models in terms of both point forecast and prediction interval.     

Forecasting house prices for the four census regions and the aggregate US economy in a data-rich environment

This article considers the ability of large-scale (involving 145 fundamental variables) time-series models, estimated by dynamic factor analysis and Bayesian shrinkage, to forecast real house price growth rates of the four US census regions and the aggregate US economy. Besides the standard Minnesota prior, we also use additional priors that constrain the sum of coefficients of the VAR models. We compare 1- to 24-months-ahead forecasts of the large-scale models over an out-of-sample horizon of 1995:01–2009:03, based on an in-sample of 1968:02–1994:12, relative to a random walk model, a small-scale VAR model comprising just the five real house price growth rates and a medium-scale VAR model containing 36 of the 145 fundamental variables besides the five real house price growth rates. In addition to the forecast comparison exercise across small-, medium- and large-scale models, we also look at the ability of the ‘optimal’ model (i.e. the model that produces the minimum average mean squared forecast error) for a specific region in predicting ex ante real house prices (in levels) over the period of 2009:04 till 2012:02. Factor-based models (classical or Bayesian) perform the best for the North East, Mid-West, West census regions and the aggregate US economy and equally well to a small-scale VAR for the South region. The ‘optimal’ factor models also tend to predict the downward trend in the data when we conduct an ex ante forecasting exercise. Our results highlight the importance of information content in large number of fundamentals in predicting house prices accurately.     

The effects of monetary policy using structural factor analysis

The traditional Vector Autoregression (VAR) method is widely used to trace out the effects of monetary policy innovations on the economy. However, this method suffers from the curse of dimensionality, so that in practice VARs are estimated on a limited number of variables, leading to a potential missing information problem. In this article we use the method of structural factor analysis to evaluate the effects of monetary policy on key macroeconomic variables in a data rich environment. This methodology allows us to extract information on monetary policy and its impact on the economy from a much larger data set than is possible with the traditional VAR method. We propose two structural factor models. One is the Structural Factor Augmented Vector Autoregressive (SFAVAR) model and the other is the Structural Factor Vector Autoregressive (SFVAR) model. Compared to the traditional VAR, both models incorporate information from hundreds of data series, series that can be and are monitored by the central bank in setting policy. Moreover, the factors used are structurally meaningful, a feature that adds to the understanding of the ‘black box’ of the monetary transmission mechanism. Both models generate qualitatively reasonable impulse response functions. For the SFVAR model, both the price puzzle and the liquidity puzzle are eliminated.     

Was the recent downturn in US real GDP predictable?

This article uses a small set of variables – real GDP, the inflation rate and the short-term interest rate – and a rich set of models – atheoretical (time series) and theoretical (structural), linear and nonlinear, as well as classical and Bayesian models – to consider whether we could have predicted the recent downturn of the US real GDP. Comparing the performance of the models to the benchmark random-walk model by root mean-square errors, the two structural (theoretical) models, especially the nonlinear model, perform well on average across all forecast horizons in our ex post, out-of-sample forecasts, although at specific forecast horizons certain nonlinear atheoretical models perform the best. The nonlinear theoretical model also dominates in our ex ante, out-of-sample forecast of the Great Recession, suggesting that developing forward-looking, microfounded, nonlinear, dynamic stochastic general equilibrium models of the economy may prove crucial in forecasting turning points.     

Short-term inflation forecasting models for Turkey and a forecast combination analysis

In this paper, we produce short term forecasts for the inflation in Turkey, using a large number of econometric models. In particular, we employ univariate models, decomposition based approaches (both in frequency and time domain), a Phillips curve motivated time varying parameter model, a suite of VAR and Bayesian VAR models and dynamic factor models. Our findings suggest that the models which incorporate more economic information outperform the benchmark random walk, and the relative performance of forecasts are on average 30% better for the first two quarters ahead. We further combine our forecasts by means of several weighting schemes. Results reveal that, the forecast combination leads to a reduction in forecast error compared to most of the models, although some of the individual models perform alike in certain horizons.     

Employment responses to aggregate and sectoral technology shocks

This paper investigates the responses of sectoral employment in US manufacturing to a technology shock by its type: aggregate or sectoral. In order to distinguish between aggregate and sectoral shocks, we construct independent VAR models for identifying each shock separately: a factor-augmented vector autoregression (FAVAR) for aggregate shocks and a sectoral SVAR for sectoral shocks. Our aggregate model in particular extends the conventional small-scale VAR to the FAVAR framework of Bernanke et al. (2005) so that it can address the potential bias from omitted variables. The main findings are as follows: most industries exhibit negative employment responses to an aggregate technology shock while exhibiting positive responses to a sectoral technology shock. By comparing our FAVAR framework with Chang and Hong’s (2006) small-scale VAR, we show that applying the FAVAR results in significant differences in the estimated responses to an aggregate technology shock. Real rigidities (such as slow diffusion of new technology or frictional labor reallocation), rather than nominal rigidities (such as sticky prices), are crucial in accounting for the cross-industry difference in employment responses. In particular, the slow diffusion of new technology is closely related to the sluggish response of sectoral employment.     

Measuring the output gap in Switzerland with linear opinion pools

We use the recently proposed linear opinion pool methodology of Garratt et al. (2014) to construct real-time output gap estimates for Switzerland over the out-of-sample period from 2003:Q1 to 2015:Q4. The model space consists of a large number of bivariate VAR specifications for the output gap and inflation, with each VAR specification using a different estimate of the output gap, lag order, and structural break information. We find that the linear opinion pool performs rather poorly. Real-time estimates of the output gap are no more accurate than those from some simple benchmark models, no more robust to ex post revisions than the real-time estimates of the individual univariate output gaps, and do not produce more accurate forecasts of inflation. The key driver of ‘good’ forecast performance is structural break information. Once the same structural break information is conditioned upon in all prediction models, the gain from averaging over many different pools of models that utilize various output gap estimates or lag structures in the VAR specification is of negligible magnitude.     

A copula-VAR-X approach for industrial production modelling and forecasting

World economies, and especially European ones, have become strongly interconnected in the last decade and a joint modelling is required. We propose here the use of copulae to build flexible multivariate distributions, since they allow for a rich dependence structure and more flexible marginal distributions that better fit the features of empirical data, such as leptokurtosis. We use our approach to forecast industrial production series in the core European Monetary Union (EMU) countries and we provide evidence that the copula-Vector Autoregression (VAR) model outperforms or at worst compares similarly to normal VAR models, keeping the same computational tractability of the latter approach.     

Differences in measures of the fiscal multiplier and the reduced-form vector autoregression

The literature has recently asked whether the effects of fiscal policy vary with the state of the economy (Christiano, Eichenbaum, and Rebelo 2011; Rendahl 2014; Auerbach and Gorodnichenko 2012). We study this question in the context of vector autoregression (VAR) estimation. We show formally that, if (asymptotically) the parameters of the reduced-form VAR differ, then the dynamic effects of fiscal policy differ as well, generically and for any set of identification assumptions. Thus, in theory, the econometrician can detect these differences (either across time or space) generically just by relying on reduced-form VAR estimation.     

Governments’ payment discipline: The macroeconomic impact of public payment delays and arrears

This paper considers the impact of changes in governments’ payment discipline on the private sector. We argue that increased delays in public payments can affect private sector liquidity and profits and hence ultimately economic growth. We test this prediction empirically for European Union countries using two complementary approaches. First, we use annual panel data, including a newly constructed proxy for government arrears. Using panel data techniques, including methods that allow for endogeneity, we find that payment delays and to some extent estimated arrears lead to a higher likelihood of bankruptcy, lower profits and lower economic growth. While this approach allows a broad set of variables to be included, it restricts the number of time periods. We therefore complement it with a Bayesian VAR approach on quarterly data for selected countries faced with significant payment delays. With this second approach, we also find that the likelihood of bankruptcies rises when the governments increase the average payment period.     

WHAT DO VARS TELL US ABOUT THE IMPACT OF A CREDIT SUPPLY SHOCK?

In the aftermath of the recent financial crisis, a variety of structural vector autoregression (VAR) models have been proposed to identify credit supply shocks. Using a Monte Carlo experiment, we show that the performance of these models can vary substantially, with some identification schemes producing particularly misleading results. When applied to U.S. data, the estimates from the best performing VAR models indicate, on average, that credit supply shocks that raise spreads by 10 basis points reduce GDP growth and inflation by 1% after one year. These shocks were important during the Great Recession, accounting for about half the decline in GDP growth.     

Precious metals,oil and the exchange rate: contemporaneous spillovers

We investigate the contemporaneous spillovers among precious metals, crude oil and the US$ exchange rate. We contend that conventional reduced-form vector autoregressive (VAR) models based on lead/lag relations do not fully capture the interactions among these series as these models ignore the contemporaneous effects. Using a Structural VAR model, we identify these contemporaneous spillovers, which are shown to be strong and asymmetric. We further show that not taking into consideration the contemporaneous interactions among these assets leads to inaccurate findings and inevitably to inaccurate interpretations of the causal relations among them.     

Bootstrapping Smooth Functions of Slope Parameters and Innovation Variances in VAR(∞) Models*

It is common to conduct bootstrap inference in vector autoregressive (VAR) models based on the assumption that the underlying data‐generating process is of finite‐lag order. This assumption is implausible in practice. We establish the asymptotic validity of the residual‐based bootstrap method for smooth functions of VAR slope parameters and innovation variances under the alternative assumption that a sequence of finite‐lag order VAR models is fitted to data generated by a VAR process of possibly infinite order. This class of statistics includes measures of predictability and orthogonalized impulse responses and variance decompositions. Our approach provides an alternative to the use of the asymptotic normal approximation and can be used even in the absence of closed‐form solutions for the variance of the estimator. We illustrate the practical relevance of our findings for applied work, including the evaluation of macroeconomic models.     

Explaining and forecasting bank loans. Good times and crisis

This article aims to develop a parsimonious model to explain and forecast bank loans to nonfinancial companies during calm periods as well as in situations of financial turmoil. It focuses on the French context, over a period including financial, banking and sovereign debt crises. Theoretical views and intuitions led us to gauge the marginal informational content of a large set of leading indicators in VAR and VECM models, and to investigate potential nonlinearity in credit dynamics. In accordance with firms and banks’ balance sheet effects, the growth rate of equity prices appears to be one of the most interesting leading indicator as well as a significant threshold variable for explaining regime switching. However, it appears difficult to accurately predict the right credit dynamics regimes. A simple VAR model finally performs better.     

Real business cycles, sticky wages or sticky prices? The impact of technology shocks on US manufacturing

In this paper, we examine empirically the predictions of a range of theoretical models which give a prominent role to technology shocks in explaining business cycles. To this end, we estimate (4-digit SIC) industry-level VAR models for US manufacturing using real output, the real wage and utilization corrected measures of technology and labor input. Our results support both sticky-wage DGE and RBC models over sticky-price DGE models. Moreover, they cast some doubt on the importance of technology shocks as propulsive mechanism for business cycles at the industry level.     

International travelling and trade: further evidence for the case of Spanish wine based on fractional vector autoregressive specifications

This article deals with the relationship between international travelling and trade. For this purpose we focus on a particular case study: the connection between the Spanish wine exports to Germany and the German travellers to Spain. Unlike previous studies we use a methodology based on fractional Vector AutoRegressive (VAR) models, which permits us to compute the impulse responses in a similar way as in the standard VAR case. The results show that the orders of integration of the two series are constrained between 0 and 1, being higher for the arrivals series than for the exports. The impulse response analysis reveals that an increase in travelling produces a positive initial impact on trade though it tends to disappear in the long run.     

The New Keynesian Phillips curve in Europe: does it fit or does it fail?

The canonical New Keynesian Phillips curve specifies inflation as the present-value of future real marginal costs. This paper exploits projections of future real marginal costs generated by VAR models to assess the model’s ability to match the behavior of actual inflation in the Euro area. The model fits the data well at first sight. A set of bias-corrected bootstrapped confidence bands, however, reveals that this result is consistent with both a well fitting and a failing model. These findings also hold for the hybrid version of the Phillips curve.     

Oil price uncertainty and the business cycle: Accounting for the influences of global supply and demand within a VAR GARCH-in-mean framework

ABSTRACT

This article reinvestigates the influence of oil price uncertainty on real economic activity in the United States using a four-variable VAR GARCH-in-mean asymmetric BEKK model. In contrast to previous studies in this area, the analysis focuses on business cycle fluctuations and we control for global supply and demand factors that might affect the real price of oil, its volatility as well as the US economy. We find that – even after accounting for these factors – oil price uncertainty still has a highly significant negative influence on the US business cycle. Our computations show that the effect is economically important during several periods, mostly after a significant variance shift in the mid-1980s. We simultaneously estimate the effect on the global business cycle but find that it is comparatively weak. Finally, significant spillover effects in the GARCH model suggest that oil price volatility is a gauge and channel of transmission of more general macroeconomic shocks and uncertainty. These linkages are particularly strong in case of unexpected bad news.     

Systemic risk contagion in FX market: A frequency connectedness and network analysis

In this study, we analyse systemic risk contagion between a set of most actively traded currencies (EURO, JPY, GBP, AUD, CAD and CHF) by application of VAR based frequency connectedness proposed by Baruník and K?ehlík. By using this novel approach, we gauge foreign exchange (FX) market connectedness in 200‐day frequency band using spectral representation of variance decompositions of VAR and identify directional spillovers between the most actively traded foreign exchange rates. Dynamics of the overall spillover index reveals that the index capture well‐known financial stress incidents properly. Finally, network topology of directional spillovers between currency pairs is provided for visulalization interconnectedness between them.