Econometric Time Series
http://lists.repec.org/mailman/listinfo/nep-ets
Econometric Time Series2014-09-05Yong YinBootstrap Score Tests for Fractional Integration in Heteroskedastic ARFIMA Models, with an Application to Price Dynamics in Commodity Spot and Futures Markets
http://d.repec.org/n?u=RePEc:aah:create:2014-22&r=ets
Empirical evidence from time series methods which assume the usual I(0)/I(1) paradigm suggests that the efficient market hypothesis, stating that spot and futures prices of a commodity should cointegrate with a unit slope on futures prices, does not hold. However, these statistical methods are known to be unreliable if the data are fractionally integrated. Moreover, spot and futures price data tend to display clear patterns of time-varying volatility which also has the potential to invalidate the use of these methods. Using new tests constructed within a more general heteroskedastic fractionally integrated model we are able to find a body of evidence in support of the efficient market hypothesis for a number of commodities. Our new tests are wild bootstrap implementations of score-based tests for the order of integration of a fractionally integrated time series. These tests are designed to be robust to both conditional and unconditional heteroskedasticity of a quite general and unknown form in the shocks. We show that the asymptotic tests do not admit pivotal asymptotic null distributions in the presence of heteroskedasticity, but that the corresponding tests based on the wild bootstrap principle do. A Monte Carlo simulation study demonstrates that very significant improvements infinite sample behaviour can be obtained by the bootstrap vis-à-vis the corresponding asymptotic tests in both heteroskedastic and homoskedastic environments.Giuseppe Cavaliere, Morten Ørregaard Nielsen, A.M. Robert Taylor2014-08-13Bootstrap, efficient market hypothesis, fractional integration, score tests, spot and futures commodity prices, time-varying volatilitySpectrum-based estimators of the bivariate Hurst exponent
http://d.repec.org/n?u=RePEc:arx:papers:1408.6637&r=ets
We introduce two new estimators of the bivariate Hurst exponent in the power-law cross-correlations setting -- the cross-periodogram and $X$-Whittle estimators. As the spectrum-based estimators are dependent on the part of the spectrum taken into consideration during estimation, a simulation study showing the performance of the estimators under varying bandwidth parameter as well as correlation between processes and their specification is provided as well. The newly introduced estimators are less biased than the already existent averaged periodogram estimator which, however, has slightly lower variance. The spectrum-based estimators can serve as a good complement to the popular time domain estimators.Ladislav Kristoufek2014-08Forecasting Exchange Rates under Model and Parameter Uncertainty
http://d.repec.org/n?u=RePEc:cqe:wpaper:3214&r=ets
We introduce a forecasting method that closely matches the econometric properties required by the theory on exchange rate prediction. Our approach formally models (i) when (and if) explanatory variables enter or leave a regression model, (ii) the degree of parameter instability, (iii) the (potentially) rapidly changing relevance of regressors, and (iv) the appropriate shrinkage intensity over time. We consider (short-term) forecasting of six major US dollar exchange rates using a standard set of macro fundamentals. Our results indicate the importance of shrinkage and flexible model selection/averaging criteria to avoid poor forecasting results.Joscha Beckmann, Rainer Schüssler2014-08Exchange rates forecasting, time-varying parameter models, shrinkage, model selection/averagingESTIMATION OF VECTOR ERROR CORRECTION MODEL WITH GARCH ERRORS: MONTE CARLO SIMULATION AND APPLICATIONS
http://d.repec.org/n?u=RePEc:ekd:006356:7002&r=ets
The standard vector error correction (VEC) model assumes the iid normal distribution of disturbance term in the model. This paper extends this assumption to include GARCH process. We call this model as VEC-GARCH model. However as the number of parameters in a VEC-GARCH model is large, the maximum likelihood (ML) method is computationally demanding. To overcome these computational difficulties, the first part of this paper searches for alternative estimation methods and compares them by Monte Carlo simulation based on a relatively small scale VEC-GARCH model; an unrestricted VECM equation system with three variables and lag of 1. After rewriting VEC-GARCH model into Seemingly Unrelated Regression (SUR) model we apply a feasible generalized least square (FGLS) estimator. As a result FGLS estimator shows comparable performance to ML estimator. Furthermore a small scale of empirical study is presented to see the applicability of the FGLS. In our simulation we found that the performance of FGLS-GARCH estimator is as good as that of MLE and both estimators are better than OLS and the standard VECM that ignore the error structure.we apply a VEC-GARCH model to real international asset pricing data and test conditional CAPM by using FGLS-GARCH estimation strategy. Since our model is relatively large; it is involving 12 stock market indexes, computational problems arise in estimating the expected returns under VEC-GARCH model and in testing the conditional CAPM by using MLE. Considering the heteroscedasticity and cross-correlation in the error terms of international stock market returns, International Capital Asset Pricing Model (CAPM) is reinvestigated under SUR with GARCH (SUR-GARCH) errors. We modified FGLS estimator to take into account multivariate GARCH error structure in estimating the model. World market portfolio was constructed to ensure that the market portfolio is mean-variance efficient under no restriction on short selling and borrowing at riskless rate. CAPM fits well only on ex-post SUR test, but it is rejected on SUR-GARCH for both ex-ante and ex-post test. However, this paper found that CAPM could be applied for most stock market indexes when each equation was analyzed individually.Kusdhianto Setiawan, Koichi Maekawa2014-07-03United States, United Kingdom, Germany, Singapore, Hong Kong, Argentina, Brazil, China, Indonesia, Malaysia, Mexico, Forecasting and projection methods, FinanceEconometric Methods for Modelling Systems with a Mixture of I(1) and I(0) Variables
http://d.repec.org/n?u=RePEc:ekd:006356:7225&r=ets
This paper considers structural models when both I(1) and I(0) variables are present. The structural shocks associated with either set of variables could be permanent or transitory. We therefore classify the shocks as (P1,P0) and (T1,T0), where P/T distinguishes permanent and transitory, while 1/0 means they are attached to either I(1) or I(0) variables. We first analyse what happens when there are P0 shocks. This is done using a sequence of examples and shows a variety of outcomes that differ from standard results in the cointegration literature. Then conditions are derived upon the nature of the SVAR in the event that T0 (and no P0) shocks are present. Following this a general method that allows for either P0 or T0 shocks is described and related to the literature that treats I(0) variables as cointegrating with themselves. Finally, we turn to an examination of a well-known empirical SVAR where there are P0 shocks. This SVAR is re-formulated so that the extra shock coming from the introduction of an I(0) variable does not affect relative prices in the long-run i.e. it is T0, and it is found that this has major implications for whether there is a price puzzle. It is also shown how to handle long-run parametric restrictions in the presence of P0 shocks when some shocks are identified using sign restrictions.Please see attachment.Please see attachment.Hyeon-Seung Huh, Lance Fisher, Adrian Pagan2014-07-03Please see attachment., Macroeconometric modeling, Macroeconometric modelingModel Uncertainty in Panel Vector Autoregressive Models
http://d.repec.org/n?u=RePEc:pra:mprapa:58131&r=ets
We develop methods for Bayesian model averaging (BMA) or selection (BMS) in Panel Vector Autoregressions (PVARs). Our approach allows us to select between or average over all possible combinations of restricted PVARs where the restrictions involve interdependencies between and heterogeneities across cross-sectional units. The resulting BMA framework can find a parsimonious PVAR specification, thus dealing with overparameterization concerns. We use these methods in an application involving the euro area sovereign debt crisis and show that our methods perform better than alternatives. Our findings contradict a simple view of the sovereign debt crisis which divides the euro zone into groups of core and peripheral countries and worries about financial contagion within the latter group.Koop, Gary, Korobilis, Dimitris2014Bayesian model averaging, stochastic search variable selection, financial contagion, sovereign debt crisisDensity Forecasting using Bayesian Global Vector Autoregressions with Common Stochastic Volatility
http://d.repec.org/n?u=RePEc:wiw:wiwwuw:wuwp179&r=ets
This paper puts forward a Bayesian Global Vector Autoregressive Model with Common Stochastic Volatility (B-GVAR-CSV). We assume that country specific volatility is driven by a single latent stochastic process, which simplifies the analysis and implies significant computational gains. Apart from computational advantages, this is also justified on the ground that the volatility of most macroeconomic quantities considered in our application tends to follow a similar pattern. Furthermore, Minnesota priors are used to introduce shrinkage to cure the curse of dimensionality. Finally, this model is then used to produce predictive densities for a set of macroeconomic aggregates. The dataset employed consists of quarterly data spanning from 1995:Q1 to 2012:Q4 and includes 45 economies plus the Euro Area. Our results indicate that stochastic volatility specifications influences accuracy along two dimensions: First, it helps to increase the overall predictive fit of our model. This result can be seen for some variables under scrutiny, most notably for real GDP and short-term interest rates. Second, it helps to make the model more resilient with respect to outliers and economic crises. This implies that when evaluated over time, the log predictive scores tend to show significantly less variation as compared to homoscedastic models.Florian Huber2014-07Density Forecasting, Stochastic Volatility, Global vector autoregressions