|
on Econometric Time Series |
| By: | Catherine Doz (PJSE - Paris Jourdan Sciences Economiques - UP1 - Université Panthéon-Sorbonne - ENS Paris - École normale supérieure - Paris - INRA - Institut National de la Recherche Agronomique - EHESS - École des hautes études en sciences sociales - ENPC - École des Ponts ParisTech - CNRS - Centre National de la Recherche Scientifique, PSE - Paris School of Economics); Peter Fuleky (University of Hawaii) |
| Abstract: | Dynamic factor models are parsimonious representations of relationships among time series variables. With the surge in data availability, they have proven to be indispensable in macroeconomic forecasting. This chapter surveys the evolution of these models from their pre-big-data origins to the large-scale models of recent years. We review the associated estimation theory, forecasting approaches, and several extensions of the basic framework. |
| Keywords: | dynamic factor models,big data,two-step estimation,time domain,frequency domain,structural breaks |
| Date: | 2019–07 |
| URL: | http://d.repec.org/n?u=RePEc:hal:psewpa:halshs-02262202&r=all |
| By: | Thomas Despois (PJSE - Paris Jourdan Sciences Economiques - UP1 - Université Panthéon-Sorbonne - ENS Paris - École normale supérieure - Paris - INRA - Institut National de la Recherche Agronomique - EHESS - École des hautes études en sciences sociales - ENPC - École des Ponts ParisTech - CNRS - Centre National de la Recherche Scientifique, PSE - Paris School of Economics); Catherine Doz (PJSE - Paris Jourdan Sciences Economiques - UP1 - Université Panthéon-Sorbonne - ENS Paris - École normale supérieure - Paris - INRA - Institut National de la Recherche Agronomique - EHESS - École des hautes études en sciences sociales - ENPC - École des Ponts ParisTech - CNRS - Centre National de la Recherche Scientifique, PSE - Paris School of Economics) |
| Abstract: | Dynamic factor models (DFMs) have been successfully used in various applications, and have become increasingly popular. Nevertheless, they suffer from two weaknesses. First, the factors are very difficult to interpret with the usual estimation methods. Secondly, there is recent and mounting evidence that they can be subject to structural instability, potentially leading to an inconsistent estimation of the factors. The central point of this paper is to tackle the uninterpretability issue. We consider two families of methods to get interpretable factors: factor rotations and sparse PCA. When using a large dataset of US macroeconomic and financial variables, they recover the same factor representation, with a simple structure in the loadings offering a clear economic interpretation of the factors, which appears to be stable over time. This provides new lens for the DFM applications, and especially to study the structural instability issue. Using this factor representation, we find no evidence for the emergence of a new factor or for a widespread break in the factor loadings at a given time. The structural instability in the model seems to rather consist in the order of the factors switching over time (apparently in relation with the structural breaks in the data), and breaks in a limited number of loadings which can be localized and interpreted. |
| Keywords: | dynamic factor models,factor rotations,sparse PCA,structural break |
| Date: | 2019–08 |
| URL: | http://d.repec.org/n?u=RePEc:hal:psewpa:halshs-02235543&r=all |
| By: | Alessio Volpicella (Queen Mary University of London) |
| Abstract: | Sign-restricted Structural Vector Autoregressions (SVARs) are increasingly common. However, they usually result in a set of structural parameters that have very different implications in terms of impulse responses, elasticities, historical decomposition and forecast error variance decomposition (FEVD). This makes it difficult to derive meaningful economic conclusions, and there is always the risk of retaining structural parameters with implausible implications. This paper imposes bounds on the FEVD as a way of sharpening set-identification induced by sign restrictions. Firstly, in a bivariate and trivariate setting, this paper analytically proves that bounds on the FEVD reduce the identified set. For higher dimensional SVARs, I establish the conditions in which the placing of bounds on the FEVD delivers a non-empty set and sharpens inference; algorithms to detect non-emptiness and reduction are also provided. Secondly, under a convexity criterion, a prior-robust approach is proposed to construct estimation and inference. Thirdly, this paper suggests a procedure to derive theory-driven bounds that are consistent with the implications of a variety of popular, but different, DSGE models, with real, nominal, and financial frictions, and with sufficiently wide ranges for their parameters. The methodology is generalized to incorporate uncertainty about the bounds themselves. Fourthly, a Monte-Carlo exercise verifies the effectiveness of those bounds in identifying the data-generating process relative to sign restrictions. Finally, a monetary policy application shows that bounds on the FEVD tend to remove unreasonable implications, increase estimation precision, sharpen and also alter the inference of models identified through sign restrictions. |
| Keywords: | Bounds, Forecast Error Variance, Monetary Policy, Set Identification, Sign Restrictions, Structural Vector Autoregressions (SVARs) |
| JEL: | C32 C53 E10 E52 |
| Date: | 2019–07–29 |
| URL: | http://d.repec.org/n?u=RePEc:qmw:qmwecw:890&r=all |
| By: | Germano Ruisi (Queen Mary University of London) |
| Abstract: | In recent years local projections have become a more and more popular methodology for the estimation of impulse responses. Besides being relatively easy to implement, the main strength of this approach relative to the traditional VAR one is that there is no need to impose any specific assumption on the dynamics of the data. This paper models local projections in a time-varying framework and provides a Gibbs sampler routine to estimate them. A simulation study shows how the performance of the algorithm is satisfactory while the usefulness of the model developed here is shown through an application to fiscal policy shocks. |
| Keywords: | Time-Varying Coefficients, Local Projections |
| JEL: | C11 C32 C36 E32 |
| Date: | 2019–07–29 |
| URL: | http://d.repec.org/n?u=RePEc:qmw:qmwecw:891&r=all |
| By: | Anne Opschoor (Vrije Universiteit Amsterdam); André Lucas (Vrije Universiteit Amsterdam) |
| Abstract: | We propose a new score-driven model to capture the time-varying volatility and tail behavior of realized kernels. We assume realized kernels follow an F distribution with two time-varying degrees-of-freedom parameters, accounting for the Vol-of-Vol and the tail shape of the realized kernel distribution. The resulting score-driven dynamics imply that the influence of large (outlying) realized kernels on future volatilities and tail-shapes is mitigated. We apply our model to 30 stocks from the S&P 500 index over the period 2001-2014. The results show that tail shapes vary over time, even after correcting for the time-varying mean and Vol-of-Vol of the realized kernels. The model outperforms a number of recent competitors, both in-sample and out-of-sample. In particular, accounting for time-varying tail shapes matters for both density forecasts and forecasts of volatility risk quantiles. |
| Keywords: | realized kernel, heavy tails, F distribution, time-varying shape-parameter, Vol-of-Vol, score-driven dynamics |
| JEL: | C32 C58 |
| Date: | 2019–07–31 |
| URL: | http://d.repec.org/n?u=RePEc:tin:wpaper:20190051&r=all |
| By: | McCracken, Michael W. (Federal Reserve Bank of St. Louis) |
| Abstract: | We investigate a test of equal predictive ability delineated in Giacomini and White (2006; Econometrica). In contrast to a claim made in the paper, we show that their test statistic need not be asymptotically Normal when a fixed window of observations is used to estimate model parameters. An example is provided in which, instead, the test statistic diverges with probability one under the null. Simulations reinforce our analytical results. |
| Keywords: | prediction; out-of-sample; inference |
| JEL: | C12 C52 C53 |
| Date: | 2019–07–29 |
| URL: | http://d.repec.org/n?u=RePEc:fip:fedlwp:2019-018&r=all |
| By: | Jérôme TRINH (Institut Polytechnique de Paris, CREST; Thema, University of Cergy-Pontoise.) |
| Abstract: | his article develops a methodology to compute up-to-date quarterly macroeconomic data for emerging countries by adapting a well known method of temporal disaggregation to time series with small sample size and instable relationships between them. By incorporating di erent procedures of structural break detection, the prediction of higher-frequency estimations of yearly oficial data can be improved. A methodology with a model selection procedure and disaggregation formulas is proposed. Its predictive performance is assessed by using empirical advanced countries data and simulated time series. An application to the Chinese national accounts allows the estimation of the cyclical components of the Chinese expenditure accounts and shows the Chinese economy to have second order moments more in line with emerging countries than advanced economies like the United States. |
| Keywords: | Time series, macroeconomic forecasting, disaggregation, structural change, business cycles, emerging economies, |
| Date: | 2019–06–27 |
| URL: | http://d.repec.org/n?u=RePEc:crs:wpaper:2019-11&r=all |
| By: | Thomas Chalaux; Yvan Guillemette |
| Abstract: | This paper describes the methodology used in the OECD Economics Department to produce historical estimates and short-run projections of potential output. These estimates are used mainly in the OECD Economic Outlook, in country surveys and as starting point for long-run scenarios. Total-economy potential output is modelled using a constant-returns-to-scale Cobb-Douglas production function with fixed factor shares. The three main inputs are labour, fixed capital excluding housing and labour efficiency, the latter obtained as a decomposition residual. The trend unemployment rate is estimated by Kalman filtering within a forward-looking Phillips curve. Other trend components are obtained by HP-filtering but labour efficiency and the labour force participation rate are cyclically adjusted before filtering to help alleviate the end-point problem associated with filters. This pre-filtering cyclical adjustment is especially helpful at cyclical turning points. It helps to lower the cyclicality of potential output as well as the extent of future revisions. |
| Keywords: | capital stock, labour efficiency, NAIRU, output gap, potential growth, potential output |
| JEL: | E20 E32 |
| Date: | 2019–08–05 |
| URL: | http://d.repec.org/n?u=RePEc:oec:ecoaaa:1563-en&r=all |
| By: | Cuaresma, Jesús Crespo; Huber, Florian; Onorante, Luca |
| Abstract: | This paper proposes a large-scale Bayesian vector autoregression with factor stochastic volatility to investigate the macroeconomic consequences of international uncertainty shocks in G7 countries. The curse of dimensionality is addressed by means of a global-local shrinkage prior that mimics certain features of the well-known Minnesota prior, yet provides additional flexibility in terms of achieving shrinkage. The factor structure enables us to identify an international uncertainty shock by assuming that it is the joint volatility process that determines the dynamics of the variance-covariance matrix of the common factors. To allow for first and second moment shocks we, moreover, assume that the uncertainty factor enters the VAR equation as an additional regressor. Our findings suggest that the estimated uncertainty measure is strongly connected to global equity price volatility, closely tracking other prominent measures commonly adopted to assess uncertainty. The dynamic responses of a set of macroeconomic and financial variables show that an international uncertainty shock exerts large effects on all economies and variables under consideration. JEL Classification: C30, E52, F41, E32 |
| Keywords: | factor stochastic volatility, global propagation of shocks, global uncertainty, vector autoregressive models |
| Date: | 2019–07 |
| URL: | http://d.repec.org/n?u=RePEc:ecb:ecbwps:20192302&r=all |
| By: | Strohsal, Till; Wolf, Elias |
| Abstract: | Data revisions to national accounts pose a serious challenge to policy decision making. Well-behaved revisions should be unbiased, small and unpredictable. This paper shows that revisions to German national accounts are biased, large and predictable. Moreover, using filtering techniques designed to process data subject to revisions, the real-time forecasting performance of initial releases can be increased by up to 17%. For total real GDP growth, however, the initial release is an optimal forecast. Yet, given the results for disaggregated variables, the averaging-out of biases and inefficiencies at the aggregate GDP level appears to be good luck rather than good forecasting. |
| Keywords: | Revisions,Real-Time Data,German National Accounts,Nowcasting |
| JEL: | C22 C53 C82 E66 |
| Date: | 2019 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:fubsbe:201911&r=all |