|
on Econometric Time Series |
|
Issue of 2020‒01‒06
eleven papers chosen by Jaqueson K. Galimberti Auckland University of Technology |
| By: | Xiaohong Chen (Cowles Foundation, Yale University); Zhuo Huang (Peking University); Yanping Yi (School of Economics and Academy of Financial Research, Zhejiang University) |
| Abstract: | This paper considers estimation of semi-nonparametric GARCH ï¬ ltered copula models in which the individual time series are modelled by semi-nonparametric GARCH and the joint distributions of the multivariate standardized innovations are characterized by parametric copulas with nonparametric marginal distributions. The models extend those of Chen and Fan (2006) to allow for semi-nonparametric conditional means and volatilities, which are estimated via the method of sieves such as splines. The ï¬ tted residuals are then used to estimate the copula parameters and the marginal densities of the standardized innovations jointly via the sieve maximum likelihood (SML). We show that, even using nonparametrically ï¬ ltered data, both our SML and the two-step copula estimator of Chen and Fan (2006) are still root-n consistent and asymptotically normal, and the asymptotic variances of both estimators do not depend on the nonparametric ï¬ ltering errors. Even more surprisingly, our SML copula estimator using the ï¬ ltered data achieves the full semiparametric efficiency bound as if the standardized innovations were directly observed. These nice properties lead to simple and more accurate estimation of Value-at-Risk (VaR) for multivariate ï¬ nancial data with flexible dynamics, contemporaneous tail dependence and asymmetric distributions of innovations. Monte Carlo studies demonstrate that our SML estimators of the copula parameters and the marginal distributions of the standardized innovations have smaller variances and smaller mean squared errors compared to those of the two-step estimators in ï¬ nite samples. A real data application is presented. |
| Keywords: | Semi-nonparametric dynamic models, Residual copulas, Semiparametric multistep, Residual sieve maximum likelihood, Semiparametric efficiency |
| JEL: | C14 C22 G32 |
| Date: | 2019–10 |
| URL: | http://d.repec.org/n?u=RePEc:cwl:cwldpp:2215&r=all |
| By: | Jörg Breitung (Institute of Econometrics, University of Cologne); Ralf Brüggemann (Department of Economics, University of Konstanz) |
| Abstract: | In this paper we provide a general framework for linear projection estimators for impulse responses in structural vector autoregressions (SVAR). An important advantage of our projection estimator is that for a large class of SVAR systems (that includes the recursive (Cholesky) identification scheme) standard OLS inference is valid without adjustment for generated regressors, autocorrelated errors or nonstationary variables. We also provide a framework for SVAR models that can be estimated by instrumental (proxy) variables. We show that this class of models (that includes also identification by long-run restrictions) result in a set of quadratic moment conditions that can be used to obtain the asymptotic distribution of this estimator, whereas standard inference based on instrumental variable (IV) projections is invalid. Furthermore, we propose a generalized least squares (GLS) version of the projections that performs similarly to the conventional (iterated) method of estimating impulse responses by inverting the estimated SVAR representation into the MA(∞) representation. Monte Carlo experiments indicate that the proposed OLS projections perform similarly to Jord`a’s (2005) projection estimator but enables us to apply standard inference on the estimated impulse responses. The GLS versions of the projections provide estimates with much smaller standard errors and confidence intervals whenever the horizon h of the impulse responses gets large. |
| Keywords: | structural vector autoregressive models, impulse responses, local projections |
| JEL: | C32 C51 |
| Date: | 2019–12–11 |
| URL: | http://d.repec.org/n?u=RePEc:knz:dpteco:1905&r=all |
| By: | Sebastian Ankargren; Paulina Jon\'eus |
| Abstract: | We discuss the issue of estimating large-scale vector autoregressive (VAR) models with stochastic volatility in real-time situations where data are sampled at different frequencies. In the case of a large VAR with stochastic volatility, the mixed-frequency data warrant an additional step in the already computationally challenging Markov Chain Monte Carlo algorithm used to sample from the posterior distribution of the parameters. We suggest the use of a factor stochastic volatility model to capture a time-varying error covariance structure. Because the factor stochastic volatility model renders the equations of the VAR conditionally independent, settling for this particular stochastic volatility model comes with major computational benefits. First, we are able to improve upon the mixed-frequency simulation smoothing step by leveraging a univariate and adaptive filtering algorithm. Second, the regression parameters can be sampled equation-by-equation in parallel. These computational features of the model alleviate the computational burden and make it possible to move the mixed-frequency VAR to the high-dimensional regime. We illustrate the model by an application to US data using our mixed-frequency VAR with 20, 34 and 119 variables. |
| Date: | 2019–12 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1912.02231&r=all |
| By: | Carlos Cesar Trucios-Maza; João H. G Mazzeu; Luis K. Hotta; Pedro L. Valls Pereira; Marc Hallin |
| Abstract: | General dynamic factor models have demonstrated their capacity to circumvent the curse of dimensionality in time series and have been successfully applied in many economic and financial applications. However, their performance in the presence of outliers has not been analysed yet. In this paper, we study the impact of additive outliers on the identification, estimation and forecasting performance of general dynamic factor models. Based on our findings, we propose robust identification, estimation and forecasting procedures. Our proposal is evaluated via Monte Carlo experiments and in empirical data. |
| Keywords: | Dimension reduction; Forecast; Jumps; Large panels |
| Date: | 2019–12 |
| URL: | http://d.repec.org/n?u=RePEc:eca:wpaper:2013/298201&r=all |
| By: | Mario Alloza (Banco de España); Jesús Gonzalo (Universidad Carlos III de Madrid); Carlos Sanz (Banco de España) |
| Abstract: | We show that several shocks identified without restrictions from a model, and frequently used in the empirical literature, display some persistence. We demonstrate that the two leading methods to recover impulse responses to shocks (moving average representations and local projections) treat persistence differently, hence identifying different objects. In particular, standard local projections identify responses that include an effect due to the persistence of the shock, while moving average representations implicitly account for it. We propose methods to re-establish the equivalence between local projections and moving average representations. In particular, the inclusion of leads of the shock in local projections allows to control for its persistence and renders the resulting responses equivalent to those associated to counterfactual non-serially correlated shocks. We apply this method to well-known empirical work on fiscal and monetary policy and find that accounting for persistence has a sizable impact on the estimates of dynamic effects. |
| Keywords: | impulse response function, local projection, shock, fiscal policy, monetary policy |
| JEL: | C32 E32 E52 E62 |
| Date: | 2019–12 |
| URL: | http://d.repec.org/n?u=RePEc:bde:wpaper:1944&r=all |
| By: | Anisha Ghosh (Institute for Fiscal Studies); Oliver Linton (Institute for Fiscal Studies and University of Cambridge) |
| Abstract: | We propose a solution to the measurement error problem that plagues the estimation of the relation between the expected return of the stock market and its conditional variance due to the latency of these conditional moments. We use intra-period returns to construct a nonparametric proxy for the latent conditional variance in the first step which is subsequently used as an input in the second step to estimate the parameters characterizing the risk-return tradeoff via a GMM approach. We propose a bias-correction to the standard GMM estimator derived under a double asymptotic framework, wherein the number of intra-period returns, N, as well as the number of low frequency time periods, T , simultaneously go to infinity. Simulation exercises show that the bias-correction is particularly relevant for small values of N which is the case in empirically realistic scenarios. The methodology lends itself to additional applications, such as the empirical evaluation of factor models, wherein the factor betas may be estimated using intra-period returns and the unexplained returns or alphas subsequently recovered at lower frequencies. |
| Date: | 2019–11–29 |
| URL: | http://d.repec.org/n?u=RePEc:ifs:cemmap:65/19&r=all |
| By: | Canova, Fabio (Norwegian Business School, CAMP and CEPR); Ferroni, Filippo (Federal Reserve Bank of Chicago) |
| Abstract: | We study what happens to identified shocks and to dynamic responses when the data generating process features q disturbances but less than q variables are used in the empirical model. Identified shocks are mongrels: they are linear combinations of current and past values of all structural disturbances and do not necessarily combine disturbances of the same type. Sound restrictions may be insufficient to obtain structural dynamics. The theory used to interpret the data and the disturbances it features determine whether an empirical model is too small. An example shows the magnitude of the distortions and the steps needed to reduce them. We revisit the evidence regarding the transmission of house price and of uncertainty shocks. |
| Keywords: | Deformation; state variables; dynamic responses; structural models; house price shocks; uncertainty shocks |
| JEL: | C31 E27 E32 |
| Date: | 2019–08–01 |
| URL: | http://d.repec.org/n?u=RePEc:hhs:rbnkwp:0378&r=all |
| By: | Luis Eduardo Castillo (Central Reserve Bank of Peru); David Florián Hoyle (Central Reserve Bank of Peru) |
| Abstract: | This paper uses a calibrated version of the Quarterly Projection Model (MPT, for its acronym in Spanish), a semi-structural dynamic model used by the Central Reserve Bank of Peru for forecasting and policy scenario analysis, to jointly estimate the output gap, potential output growth and natural interest rate of the Peruvian economy during the inflation targeting regime (between 2002 and 2017). The model is employed as a multivariate filter with a sophisticated economic structure that allows us to infer the dynamics of non-observable variables from the information provided by other variables defined ex-ante as observable. As the results from the Kalman filter are sensible to the variables declared as observable, we use five groups of variables to be defined as such to build probable ranges for our estimates. The results indicate that the estimated output gap is large in amplitude and highly persistent while potential output growth is very smooth. Therefore, most of the variation in Peruvian economic activity during the inflation targeting regime can be attributed to the former. The estimation of the output gap also proves that monetary policy has been extensively responsive to this leading indicator of inflation. Meanwhile, the real natural interest rate is estimated to be considerable stable, averaging 1,6 percent in the sample with only a sharp decline to 1,3 percent during the financial crisis. The main finding of the paper, however, is that there has been a steady deceleration of potential output growth since 2012. A growth-accounting exercise proves that this trend follows mostly a reduction in total factor productivity (TFP) growth during the same time frame (although the drop of capital and labour contributions jointly explain almost a third part of average potential output growth slowdown between 2010-2013 and 2014-2017). |
| Keywords: | Potential output, Output gap, Natural Interest Rate, Kalman Filter, Peru |
| JEL: | C51 E32 E52 |
| Date: | 2019–12 |
| URL: | http://d.repec.org/n?u=RePEc:apc:wpaper:159&r=all |
| By: | Becher, Georg |
| Abstract: | The objective of the EU Timber Regulation (EUTR), enforced since March 2013, is for importers and exporters to commit to reducing the risk of trading timber products from illegal sources in the EU. EUROSTAT time series on monthly trade with wood products from January 1988 to August 2016 were used to monitor the law’s impact. The time series, subdivided into sections before and after the implementation of EUTR, were investigated in time and frequency domains. The analyses in the time domain indicated the adequateness of the AR (1) and ARMA (1, 1) models. As the confidence intervals for their estimates before and after EUTR do not overlap, the respective time series are considered as different and the influence of EUTR legislation probable (also confirmed by the significant models with EUTR as intervening event). Long term variation of the monthly time series (March 2013 to August 2016) show an increasing linear trend for all wood products and for wood products with tropical woods excluded. Since EU imports of tropical wood were falling before EUTR, the stagnant imports thereafter are judged as uncertainty and time the markets need to adapt to a new legislative situation. The analyses in frequency domain based on inference from periodogram revealed cycles of 3, 4, 6 and 12 months, except for time series of tropical wood imports after EUTR. If cycles are thought of as inherent to import time series, this lack in tropical wood imports can be an indication of a ‘wait-and-see’ attitude of importers as a consequence of EUTR. |
| Keywords: | Research Methods/ Statistical Methods |
| Date: | 2019–12–18 |
| URL: | http://d.repec.org/n?u=RePEc:ags:jhimwp:298443&r=all |
| By: | Aminu, Alarudeen; Raifu, Isiaka Akande |
| Abstract: | The incessant fiscal deficit being experienced in different countries across the world has raised concerns about the ability of government to properly manage its revenues and expenditures. This has necessitated a flurry of studies on the relationship between government revenues and government expenditures over time. However, empirical evidence appears to be mixed, even within a country, depending on the methodological approaches adopted by each researcher. In the light of this, this study examines the asymmetric causality and cointegration between revenues and expenditures using aggregated and disaggregated data. The results of linear causality tests of Granger (1969) and Toda-Yamamoto (1995) support fiscal synchronisation hypothesis while those of nonlinear causality test of Diks and Panchenko (2006) support revenue-spending hypothesis. The results further show the existence of asymmetric cointegration between revenues and expenditures in the short-run and the long-run. The final results obtained from the decomposition of revenues into the positive and negative components show that positive change in revenues has a positive effect on expenditures and vice versa for a negative change in revenues. Based on these findings, the panacea proposed to over-reliance in revenues, particularly oil revenues as a determinant of government expenditures, is the proper management of oil revenues and other sources of revenues. The government would also need to diversify the economy so that more revenues could be available to it from other sources to finance its expenditures. |
| Keywords: | Government Revenues, Government Expenditures, Asymmetries, Causality, Cointegration |
| JEL: | C20 C50 H27 |
| Date: | 2018–12–12 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:97880&r=all |
| By: | ŞENTÜRK, İsmail; Ali, Amjad |
| Abstract: | This paper has tried to analyze the socioeconomic determinants of total as well as gender specific life expectancy in Turkey from 1971 to 2017. Data stationarity has been checked by ADF, PP and DFGLS unit root tests and cointegration has been checked with the help of the ARDL bound testing method. The estimated results show that the overall level of education, purchasing power and economic development have a significant role in deciding total average life expectancy in Turkey. Whereas, population growth and environmental degradation have an insignificant contribution in deciding total average life expectancy in Turkey. Estimates show environmental degradation, purchasing power and level of male education have contributed significantly in male life expectancy in Turkey. Economic development and share of the male population have an insignificant role in deciding life expectancy of male in Turkey. Environmental degradation, the level of female education, fertility rates and female population significantly effected female life expectancy, but purchasing power has an insignificant role in deciding life expectancy of female in Turkey. The results recommend that the government of Turkey should enhance the level of education and try to stable purchasing power and sustainable development with controlled fertility rates for higher level life expectancy. |
| Keywords: | life expectancy, education, environmental degradation, population growth |
| JEL: | O1 Q0 |
| Date: | 2019 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:97815&r=all |