|
on Econometric Time Series |
| By: | Matt P. Dziubinski (Aarhus University and CREATES) |
| Abstract: | We present and evaluate a numerical optimization method (together with an algorithm for choosing the starting values) pertinent to the constrained optimization problem arising in the estimation of the GARCH models with inequality constraints, in particular the Simplied Component GARCH Model (SCGARCH), together with algorithms for the objective function and analytical gradient computation for SCGARCH. |
| Keywords: | Constrained optimization, GARCH, infeasibility, inference under constraints, nonlinear programming, performance of numerical algorithms, SCGARCH, sequential quadratic programming |
| JEL: | C32 C51 C58 C61 C63 C88 |
| Date: | 2012–01–25 |
| URL: | http://d.repec.org/n?u=RePEc:aah:create:2012-03&r=ets |
| By: | Tara M. Sinclair (George Washington University) |
| Abstract: | This article provides a discussion of Clements and Galvão’s “Forecasting with Vector Autoregressive Models of Data Vintages: US output growth and inflation.” Clements and Galvão argue that a multiple-vintage VAR model can be useful for forecasting data that are subject to revisions. Clements and Galvão draw a “distinction between forecasting future observations and revisions to past data,” which brings yet another real time data issue to the attention of forecasters. This comment discusses the importance of taking data revisions into consideration and compares the multiple-vintage VAR approach of Clements and Galvão to a state-space approach. |
| Keywords: | Real time data, Evaluating forecasts, Forecasting practice, Time series, Econometric models |
| JEL: | C53 |
| Date: | 2012–01 |
| URL: | http://d.repec.org/n?u=RePEc:gwc:wpaper:2012-001&r=ets |
| By: | Fernández Macho, Francisco Javier |
| Abstract: | We propose a two-dimensional Kalman filter approach that, additional to the information contained in futures prices evolution over time, makes use of information contained in the term structure of commodity futures along a second dimension of maturities. This time-maturity surface reflects a complete realization of the stochastic process as an alternative to standard Kalman filtering of a limited vector of futures prices along the one-dimensional time line. Thus, the proposed methodology may use the full information from the entire surface dynamics, including links from all available maturities per period, which eventually should lead to more accurate model parameter estimates. The technique is illustrated using coal futures prices. |
| Keywords: | commodity prices, two-dimensional Kalman filter, spatial analysis, energy markets, futures markets, stochastic dynamic model, |
| Date: | 2011–09 |
| URL: | http://d.repec.org/n?u=RePEc:ehu:biltok:5503&r=ets |
| By: | Díaz-Emparanza Herrero, Ignacio |
| Abstract: | When working with time series data observed at intervals smaller than a year, it is often necessary to test for the presence of seasonal unit roots. One of the most widely used methods for testing seasonal unit roots is that of HEGY, which provides test statistics with non-standard distributions. This paper describes a generalisation of this method for any periodicity and uses a response surface regressions approach to calculate the critical values and P values of the HEGY statistics whatever the periodicity and sample size of the data. The algorithms are prepared with the Gretl open source econometrics package and some new tables of critical values for daily, hourly and half-hourly data are presented. |
| Keywords: | seasonality, unit roots, surface response analysis, |
| Date: | 2011–12 |
| URL: | http://d.repec.org/n?u=RePEc:ehu:biltok:5568&r=ets |
| By: | Jozef Barunik; Ladislav Kristoufek |
| Abstract: | In this paper, we show how the sampling properties of the Hurst exponent methods of estimation change with the presence of heavy tails. We run extensive Monte Carlo simulations to find out how rescaled range analysis (R/S), multifractal detrended fluctuation analysis (MF-DFA), detrending moving average (DMA) and generalized Hurst exponent approach (GHE) estimate Hurst exponent on independent series with different heavy tails. For this purpose, we generate independent random series from stable distribution with stability exponent {\alpha} changing from 1.1 (heaviest tails) to 2 (Gaussian normal distribution) and we estimate the Hurst exponent using the different methods. R/S and GHE prove to be robust to heavy tails in the underlying process. GHE provides the lowest variance and bias in comparison to the other methods regardless the presence of heavy tails in data and sample size. Utilizing this result, we apply a novel approach of the intraday time-dependent Hurst exponent and we estimate the Hurst exponent on high frequency data for each trading day separately. We obtain Hurst exponents for S&P500 index for the period beginning with year 1983 and ending by November 2009 and we discuss the surprising result which uncovers how the market's behavior changed over this long period. |
| Date: | 2012–01 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1201.4786&r=ets |
| By: | Jozef Barunik; Lukas Vacha |
| Abstract: | In this paper we propose a new approach to estimation of the tail exponent in financial stock markets. We begin the study with the finite sample behavior of the Hill estimator under {\alpha}-stable distributions. Using large Monte Carlo simulations, we show that the Hill estimator overestimates the true tail exponent and can hardly be used on samples with small length. Utilizing our results, we introduce a Monte Carlo-based method of estimation for the tail exponent. Our proposed method is not sensitive to the choice of tail size and works well also on small data samples. The new estimator also gives unbiased results with symmetrical confidence intervals. Finally, we demonstrate the power of our estimator on the international world stock market indices. On the two separate periods of 2002-2005 and 2006-2009, we estimate the tail exponent. |
| Date: | 2012–01 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1201.4781&r=ets |
| By: | Leonidas Sandoval Junior |
| Abstract: | Financial markets worldwide do not have the same working hours. As a consequence, the study of correlation or causality between financial market indices becomes dependent on wether we should consider in computations of correlation matrices all indices in the same day or lagged indices. The answer is that we should consider both. |
| Date: | 2012–01 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1201.4586&r=ets |