|
on Econometric Time Series |
| By: | Francis X. Diebold (Department of Economics, University of Pennsylvania); Minchul Shin (Department of Economics, University of Illinois) |
| Abstract: | Despite the clear success of forecast combination in many economic environments, several important issues remain incompletely resolved. The issues relate to selection of the set of forecasts to combine, and whether some form of additional regularization (e.g., shrinkage) is desirable. Against this background, and also considering the frequently-found superiority of simple-average combinations, we propose LASSO-based procedures that select and shrink toward equal combining weights. We then provide an empirical assessment of the performance of our "egalitarian LASSO" procedures. The results indicate that simple averages are highly competitive, and that although out-of-sample RMSE improvements on simple averages are possible in principle using our methods, they are hard to achieve in real time, due to the intrinsic difficulty of small-sample real-time cross validation of the LASSO tuning parameter. We therefore propose alternative direct combination procedures, most notably "best average" combination, motivated by the structure of egalitarian LASSO and the lessons learned, which do not require choice of a tuning parameter yet outperform simple averages. |
| Keywords: | Forecast combination, forecast surveys, shrinkage, model selection, LASSO, regularization |
| JEL: | C53 |
| Date: | 2017–08–20 |
| URL: | http://d.repec.org/n?u=RePEc:pen:papers:17-017&r=ets |
| By: | Markus Heinrich; Magnus Reif |
| Abstract: | We extend the literature on economic forecasting by constructing a mixed-frequency time-varying parameter vector autoregression with stochastic volatility (MF-TVP-SVVAR). The latter is able to cope with structural changes and can handle indicators sampled at different frequencies. We conduct a real-time forecast exercise to predict US key macroeconomic variables and compare the predictions of the MF-TVP-SV-VAR with several linear, nonlinear, mixed-frequency, and quarterly-frequency VARs. Our key finding is that the MF-TVPSV-VAR delivers very accurate forecasts and, on average, outperforms its competitors. In particular, inflation forecasts benefit from this new forecasting approach. Finally, we assess the models’ performance during the Great Recession and find that the combination of stochastic volatility, time-varying parameters, and mixed-frequencies generates very precise inflation forecasts. |
| Keywords: | Time-varying parameters, forecasting, mixed-frequency models, Bayesian methods |
| JEL: | C11 C53 E32 |
| Date: | 2018 |
| URL: | http://d.repec.org/n?u=RePEc:ces:ifowps:_273&r=ets |
| By: | Altug, Sumru G.; Cakmakli, Cem; Demircan, Hamza |
| Abstract: | This paper puts forward a unified framework for the joint estimation of the indexes that can broadly capture economic and financiall conditions together with their cyclical regimes of recession and expansion. We do this by utilizing a dynamic factor model together with Markov regime switching dynamics of model parameters that specifically exploit the temporal link between the cyclical behavior of economic and financial factors. This is achieved by constructing the cycle in the financial factor using the cycle in the economic factor together with phase shifts. The resulting framework allows the financial cycle to potentially lead/lag the business cycle in a systematic manner and exploits the information in economic and financial variables for estimation of both economic and financial conditions as well as their cyclical behavior in an efficient way. We examine the potential of the model using a mixed frequency and mixed time span ragged-edge dataset for Turkey. Comparison of our framework with more conventional polar cases imposing a single common cyclical dynamics as well as independent cyclical dynamics for economic and financial conditions reveal that the proposedspecification provides precise estimates of economic and financial conditions and it delivers quite accurate probabilities of recessions that match with stylized facts. We further conduct a recursive real-time exercise of nowcasting and forecasting business cycle turning points. The results show convincing evidence of superior predictive power of our specification by signaling oncoming recessions (expansions) as early as 3.5 (3.4) months ahead of the actual realization. |
| Keywords: | Bayesian inference; Business cycle; Coincident economic index; Dynamic factor model; Financial conditions index; Markov switching |
| Date: | 2018–09 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:13171&r=ets |
| By: | Giuseppe Cavaliere (Department of Economics, University of Bologna, Italy); Heino Bohn Nielsen (Department of Economics, University of Copenhagen, Denmark); Rasmus Søndergaard Pedersen (Department of Economics, University of Copenhagen, Denmark); Anders Rahbek (Department of Economics, University of Copenhagen, Denmark) |
| Abstract: | It is a well-established fact that testing a null hypothesis on the boundary of the parameter space, with an unknown number of nuisance parameters at the boundary, is infeasible in practice in the sense that limiting distributions of standard test statistics are non-pivotal. In particular, likelihood ratio statistics have limiting distributions which can be characterized in terms of quadratic forms minimized over cones, where the shape of the cones depends on the unknown location of the (possibly mulitiple) model parameters not restricted by the null hypothesis. We propose to solve this inference problem by a novel bootstrap, which we show to be valid under general conditions, irrespective of the presence of (unknown) nuisance parameters on the boundary. That is, the new bootstrap replicates the unknown limiting distribution of the likelihood ratio statistic under the null hypothesis and is bounded (in probability) under the alternative. The new bootstrap approach, which is very simple to implement, is based on shrinkage of the parameter estimates used to generate the bootstrap sample toward the boundary of the parameter space at an appropriate rate. As an application of our general theory, we treat the problem of inference in ?nite-order ARCH models with coefficients subject to inequality constraints. Extensive Monte Carlo simulations illustrate that the proposed bootstrap has attractive ?nite sample properties both under the null and under the alternative hypothesis. |
| Keywords: | Inference on the boundary, Nuisance parameters on the boundary, ARCH models, Bootstrap |
| JEL: | C12 C22 |
| Date: | 2018–11–12 |
| URL: | http://d.repec.org/n?u=RePEc:kud:kuiedp:1810&r=ets |
| By: | Xiaowei Zhang; Peter W. Glynn |
| Abstract: | Affine jump-diffusions constitute a large class of continuous-time stochastic models that are particularly popular in finance and economics due to their analytical tractability. Methods for parameter estimation for such processes require ergodicity in order establish consistency and asymptotic normality of the associated estimators. In this paper, we develop stochastic stability conditions for affine jump-diffusions, thereby providing the needed large-sample theoretical support for estimating such processes. We establish ergodicity for such models by imposing a `strong mean reversion' condition and a mild condition on the distribution of the jumps, i.e. the finiteness of a logarithmic moment. Exponential ergodicity holds if the jumps have a finite moment of a positive order. In addition, we prove strong laws of large numbers and functional central limit theorems for additive functionals for this class of models. |
| Date: | 2018–10 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1811.00122&r=ets |
| By: | Francis Vitek |
| Abstract: | This paper investigates the sources of macrofinancial fluctuations and turbulence within the framework of an approximate linear dynamic stochastic general equilibrium model of the world economy, augmented with structural shocks exhibiting potentially asymmetric generalized autoregressive conditional heteroskedasticity. Very strong evidence of asymmetric autoregressive conditional heteroskedasticity is found, providing a basis for jointly decomposing the levels and volatilities of key macrofinancial variables into time varying contributions from sets of shocks. Risk premia shocks are estimated to contribute disproportionately to cyclical output fluctuations and turbulence during swings in financial conditions, across the fifteen largest national economies in the world. |
| Date: | 2018–11–08 |
| URL: | http://d.repec.org/n?u=RePEc:imf:imfwpa:18/238&r=ets |
| By: | Paulo M.D.C. Parente; Richard J. Smith |
| Abstract: | This paper applies a novel bootstrap method, the kernelblockbootstrap, to quasi-maximum likelihood estimation of dynamic models with stationary strong mixing data. The method rst kernel weights the components comprising the quasi-log likelihood function in an appropriate way and then samples the resultant transformed components using the standard "m out of n"bootstrap. We investigate the first order asymptotic properties of the KBB method for quasi-maximum likelihood demonstrating, in particular, its consistency and the rst-order asymptotic validity of the bootstrap approximation to the distribution of the quasi-maximum likelihood estimator. A set of simulation experiments for the mean regression model illustrates the efficacy of the kernel block bootstrap for quasi-maximum likelihood estimation. |
| Keywords: | Bootstrap; heteroskedastic and autocorrelation consistent inference; quasi-maximum likelihood estimation. |
| JEL: | C14 C15 C22 |
| Date: | 2018–11 |
| URL: | http://d.repec.org/n?u=RePEc:ise:remwps:wp0592018&r=ets |
| By: | Peter K. Friz; Paul Gassiat; Paolo Pigato |
| Abstract: | We present a new methodology to analyze large classes of (classical and rough) stochastic volatility models, with special regard to short-time and small noise formulae for option prices. Our main tool is the theory of regularity structures, which we use in the form of [Bayer et al; A regularity structure for rough volatility, 2017]. In essence, we implement a Laplace method on the space of models (in the sense of Hairer), which generalizes classical works of Azencott and Ben Arous on path space and then Aida, Inahama--Kawabi on rough path space. When applied to rough volatility models, e.g. in the setting of [Forde-Zhang, Asymptotics for rough stochastic volatility models, 2017], one obtains precise asymptotic for European options which refine known large deviation asymptotics. |
| Date: | 2018–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1811.00267&r=ets |
| By: | Juan Sebastian Cubillos-Rocha (Banco de la República de Colombia); Luis Fernando Melo-Velandia (Banco de la República de Colombia) |
| Abstract: | Panel dynamic estimators with fixed effects are biased due to the incidental parameters problem. At this regard, Hahn and Kuersteiner (2002) proposed an estimator to correct this issue. However, they only consider a panel VAR (PVAR) model with one lag. In this paper we extend this bias correction, its asymptotic and small sample properties for a more general case, a PVAR model with p lags. The simulation results indicate that the bias corrected estimator outperforms the OLS panel VAR estimator when sample size in time dimension is small, and when the persistence of the model is low. In these cases, the proposed estimator improves significantly in terms of both, the reduction of bias and mean square error. **** RESUMEN: Los estimadores de los parámetros de un modelo panel dinámico de efectos fijos son sesgados debido al problema de parámetros incidentales. Al respecto, Hahn y Kuersteiner (2002) proponen un estimador para corregir este problema. Sin embargo, ellos consideran únicamente un modelo panel VAR con un sólo un rezago. En este documento analizamos las propiedades asintóticas y de muestra pequeña del estimador corregido por sesgo para un caso más general, un modelo PVAR con p rezagos. Los resultados de las simulaciones indican que el estimador corregido por sesgo tiene un mejor desempeño con respecto al estimador panel VAR MCO cuando la dimensión temporal de la muestra (T) es pequeña, y cuando la persistencia del modelo es baja. En estos casos, el estimador propuesto presenta una disminución significativa en términos de sesgo, y de error cuadrático medio. |
| Keywords: | Panel VAR models; bias correction; restricted OLS, Modelos Panel VAR; corrección de sesgo; MCO restringido. |
| JEL: | C33 C51 C13 |
| Date: | 2018–11 |
| URL: | http://d.repec.org/n?u=RePEc:bdr:borrec:1059&r=ets |
| By: | Lui, Yiu Lim (School of Economics, Singapore Management University); Xiao, Weilin (School of Management, Zhejiang University); Yu, Jun (School of Economics, Singapore Management University) |
| Abstract: | This paper considers the grid bootstrap for constructing confidence intervals for the persistence parameter in a class of continuous time models driven by a Levy process. Its asymptotic validity is established by assuming the sampling interval (h) shrinks to zero. Its improvement over the in-fill asymptotic theory is achieved by expanding the coefficient-based statistic around its in fill asymptotic distribution which is non-pivotal and depends on the initial condition. Monte Carlo studies show that the gird bootstrap method performs better than the in-fill asymptotic theory and much better than the longspan theory. Empirical applications to U.S. interest rate data highlight differences between the bootstrap confidence intervals and the confidence intervals obtained from the in-fill and long-span asymptotic distributions. |
| Keywords: | Grid bootstrap; In-fill asymptotics; Continuous time models; Long-span asymptotics. |
| JEL: | C11 C12 |
| Date: | 2018–11–09 |
| URL: | http://d.repec.org/n?u=RePEc:ris:smuesw:2018_020&r=ets |