|
on Econometric Time Series |
| By: | George Kapetanios (King’s College, London); Massimiliano Marcellino (Bocconi University, IGIER); Fabrizio Venditti (Bank of Italy) |
| Abstract: | In this paper we introduce a non-parametric estimation method for a large Vector Autoregression (VAR) with time-varying parameters. The estimators and their asymptotic distributions are available in closed form. This makes the method computationally efficient and capable of handling information sets as large as those typically handled by factor models and Factor Augmented VARs (FAVAR). When applied to the problem of forecasting key macroeconomic variables, the method outperforms constant parameter benchmarks and large (parametric) Bayesian VARs with time-varying parameters. The tool can also be used for structural analysis. As an example, we study the time-varying effects of oil price innovations on sectoral U.S. industrial output. We find that durable consumer goods and durable materials (which together account for slightly more than one fifth of total industrial output) play a key role in explaining the changing interaction between unexpected oil price increases and U.S. business cycle fluctuations. |
| Keywords: | large VARs, time-varying parameters, non-parametric estimation, forecasting |
| JEL: | C14 C32 C53 C55 |
| Date: | 2017–06 |
| URL: | http://d.repec.org/n?u=RePEc:bdi:wptemi:td_1122_17&r=ets |
| By: | Florian Huber (Department of Economics, Vienna University of Economics and Business); Thomas Zörner (Department of Economics, Vienna University of Economics and Business) |
| Abstract: | This paper considers Bayesian estimation of the threshold vector error correction (TVECM) model in moderate to large dimensions. Using the lagged cointegrating error as a threshold variable gives rise to additional difficulties that are typically solved by relying on large sample approximations. Relying on Markov chain Monte Carlo methods we circumvent these issues by avoiding computationally prohibitive estimation strategies like the grid search. Due to the proliferation of parameters we use novel global-local shrinkage priors in the spirit of Griffin and Brown (2010). We illustrate the merits of our approach in an application to five exchange rates vis-á-vis the US dollar and assess whether a given currency is over or undervalued. Moreover, we perform a forecasting comparison to investigate whether it pays off to adopt a non-linear modeling approach relative to a set of simpler benchmark models. |
| Keywords: | non-linear modeling, shrinkage priors, multivariate cointegration, exchange rate modeling |
| JEL: | C11 C32 C53 F31 F47 |
| Date: | 2017–06 |
| URL: | http://d.repec.org/n?u=RePEc:wiw:wiwwuw:wuwp250&r=ets |
| By: | Hui, Yongchang; Wong, Wing-Keung; BAI, ZHIDONG; Zhu, Zhen-Zhen |
| Abstract: | In this paper, we propose a quick and efficient method to examine whether a time series ${X}_t$ possesses any nonlinear feature by testing a kind of dependence remained in the residuals after fitting ${X}_t$ with a linear model. The advantage of our proposed nonlinearity test is that it is not required to know the exact nonlinear features and the detailed nonlinear forms of the variable being examined. Another advantage of our proposed test is that there is no over-rejection problem which exists in some famous nonlinearity tests. Our proposed test can also be used to test whether the hypothesized model, including linear and nonlinear, to the variable being examined is appropriate as long as the residuals of the model being used can be estimated. Our simulation study shows that our proposed test is stable and powerful. We apply our proposed statistic to test whether there is any nonlinear feature in the sunspot data. The conclusion drawn from our proposed test is consistent with those from other well-established tests. |
| Keywords: | Nonlinearity, Dependence, Nonlinear test, Dependent test, Volterra expansion, Sunspots |
| JEL: | C01 C12 |
| Date: | 2017–06–13 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:79692&r=ets |
| By: | Aknouche, Abdelhakim; Bentarzi, Wissam; Demouche, Nacer |
| Abstract: | We propose a general class of non-linear mixed Poisson autoregressions whose form and parameters are periodic over time. Under a periodic contraction condition on the forms of the conditional mean, we show the existence of a unique nonanticipative solution to the model, which is strictly periodically stationary, periodically ergodic and periodically weakly dependent having in the pure Poisson case finite higher-order moments. Applications to some well-known integer-valued time series models are considered. |
| Keywords: | Periodic mixed Poisson autoregression, periodic INGARCH models, non-linear INGARCH models, weak dependence, strict periodic stationarity, periodic ergodicity, periodic contraction condition. |
| JEL: | C10 C19 C51 C62 |
| Date: | 2017–02–01 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:79650&r=ets |
| By: | Massimo Franchi ("Sapienza" University of Rome); Paolo Paruolo (European Commission, Joint Research Centre) |
| Abstract: | A generalization of the Granger and the Johansen Representation Theorems valid for any (possibly fractional) order of integration is presented. This is based on an inversion theorem that characterizes the order of the pole and the coefficients of the Laurent series representation of the inverse of a matrix function around a singular point. Explicit expressions of the matrix coecients of the (polynomial) cointegrating relations, of the common trends and of the triangular representations are provided, either starting from the Moving Average or the Auto Regressive form. This unifies the different approaches in the literature, and extends them to an arbitrary order of integration. |
| Keywords: | Cointegration, Common Trends, Triangular representation,Local Smith form, Moving Average representation, Autoregressive representation. |
| JEL: | C12 C33 C55 |
| Date: | 2017–06 |
| URL: | http://d.repec.org/n?u=RePEc:sas:wpaper:20173&r=ets |
| By: | Asai, M.; Chang, C-L.; McAleer, M.J. |
| Abstract: | The paper develops a novel realized stochastic volatility model of asset returns and realized volatility that incorporates general asymmetry and long memory (hereafter the RSV-GALM model). The contribution of the paper ties in with Robert Basmann’s seminal work in terms of the estimation of highly non-linear model specifications (“Causality tests and observationally equivalent representations of econometric models”, Journal of Econometrics, 1988), especially for specifying causal effects from returns to future volatility. This paper discusses asymptotic results of a Whittle likelihood estimator for the RSV-GALM model and a test for general asymmetry, and analyses the finite sample properties. The paper also develops an approach to obtain volatility estimates and out-of-sample forecasts. Using high frequency data for three US financial assets, the new model is estimated and evaluated. The paper compares the forecasting performance of the new model with a realized conditional volatility model. |
| Keywords: | Stochastic Volatility, Realized Measure, Long Memory, Asymmetry, Whittle likelihood, Asymptotic Distribution |
| JEL: | C13 C22 |
| Date: | 2017–04–01 |
| URL: | http://d.repec.org/n?u=RePEc:ems:eureir:100161&r=ets |