|
on Econometric Time Series |
| By: | Garroi J.-J.; Goos P.; Sörensen K. |
| Abstract: | The responses obtained from response surface designs that are run sequentially often exhibit serial correlation or time trends. The order in which the runs of the design are performed then has an impact on the precision of the parameter estimators. This article proposes the use of a variable-neighbourhood search algorithm to compute run orders that guarantee a precise estimation of the effects of the experimental factors. The importance of using good run orders is demonstrated by seeking D-optimal run orders for a central composite design in the presence of an AR(1) autocorrelation pattern. |
| Date: | 2006–10 |
| URL: | http://d.repec.org/n?u=RePEc:ant:wpaper:2006026&r=ets |
| By: | Gianluca Moretti (Banca d'Italia, Research Department) |
| Abstract: | Cointegration analysis tests for the existence of a significant long-run equilibrium among some economic variables. Standard econometric procedures to test for cointegration have proven unreliable when the long-run relation among the variables is characterized by non-linearities and persistent fluctuations around the equilibrium. As a consequence, many intuitive economic relations are empirically rejected. In this paper we propose a simple approach to account for non-linearities in the cointegrating equilibrium and possible long memory fluctuations from such equilibrium. We show that our correction allows us to test robustly for the presence of cointegration both under the null and alternative hypotheses. We apply our procedure to the Johansen-Juselius PPP-UIP database, and unlike the standard case, we do not fail to reject the null of no cointegration. |
| Keywords: | Cointegration analysis, long memory |
| JEL: | C22 C51 |
| Date: | 2007–09 |
| URL: | http://d.repec.org/n?u=RePEc:bdi:wptemi:td_642_07&r=ets |
| By: | Artis, Michael J; Clavel, Jose Garcia; Hoffmann, Mathias; Nachane, Dilip M |
| Abstract: | Strongly periodic series occur frequently in many disciplines. This paper reviews one specific approach to analyzing such series viz. the harmonic regression approach. In this paper, the five major methods suggested under this approach are critically reviewed and compared, and their empirical potential highlighted via two applications. The out-of-sample forecast comparisons are made using the Superior Predictive Ability test, which specifically guards against the perils of data snooping. Certain tentative conclusions are drawn regarding the relative forecasting ability of the different methods. |
| Keywords: | autoregressive methods; data snooping; dynamic harmonic regression; eigenvalue methods; mixed spectrum; multiple forecast comparisons |
| JEL: | C22 C53 |
| Date: | 2007–10 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:6517&r=ets |
| By: | George Kapetanios (Queen Mary, University of London) |
| Abstract: | Testing serial dependence is central to much of time series econometrics. A number of tests that have been developed and used to explore the dependence properties of various processes. This paper builds on recent work on nonparametric tests of independence. We consider a fact that characterises serially dependent processes using a generalisation of the autocorrelation function. Using this fact we build dependence tests that make use of neural network based approximations. We derive the theoretical properties of our tests and show that they have superior power properties. Our Monte Carlo evaluation supports the theoretical findings. An application to a large dataset of stock returns illustrates the usefulness of the proposed tests. |
| Keywords: | Independence, Neural networks, Strict stationarity, Bootstrap, S&P500 |
| JEL: | C32 C33 G12 |
| Date: | 2007–10 |
| URL: | http://d.repec.org/n?u=RePEc:qmw:qmwecw:wp609&r=ets |
| By: | Giacomini, Raffaella (University College London); Gottschling, Andreas (Deutsche Bank AG, Credit RiskManagement); Haefke, Christian (Department of Economics and Finance, Institute for Advanced Studies, Vienna, Austria); White, Halbert (Department of Economics, University of California, San Diego) |
| Abstract: | We explore convenient analytic properties of distributions constructed as mixtures of scaled and shifted t-distributions. A feature that makes this family particularly desirable for econometric applications is that it possesses closed-form expressions for its anti-derivatives (e.g., the cumulative density function). We illustrate the usefulness of these distributions in two applications. In the first application, we use a scaled and shifted t-distribution to produce density forecasts of U.S. inflation and show that these forecasts are more accurate, out-of-sample, than density forecasts obtained using normal or standard t-distributions. In the second application, we replicate the option-pricing exercise of Abadir and Rockinger (2003) using a mixture of scaled and shifted t-distributions and obtain comparably good results, while gaining analytical tractability. |
| Keywords: | ARMA-GARCH models, neural networks, nonparametric density estimation, forecast accuracy, option pricing, risk neutral density |
| JEL: | C63 C53 C45 |
| Date: | 2007–10 |
| URL: | http://d.repec.org/n?u=RePEc:ihs:ihsesp:216&r=ets |
| By: | Di Iorio, Francesca; Fachin, Stefano |
| Abstract: | Stability tests for cointegrating coefficients are known to have very low power with small to medium sample sizes. In this paper we propose to solve this problem by extending the tests to dependent cointegrated panels through the stationary bootstrap. Simulation evidence shows that the proposed panel tests improve considerably on asymptotic tests applied to individual series. As an empirical illustration we examined investment and saving for a panel of 14 European countries over the 1960-2002 period. While the individual stability tests, contrary to expectations and graphical evidence, in almost all cases do not reject the null of stability, the bootstrap panel tests lead to the more plausible conclusion that the long-run relationship between these two variables is likely to have undergone a break. |
| Keywords: | Panel cointegration, stationary bootstrap, parameter stability tests, FM-OLS |
| JEL: | C15 C23 |
| Date: | 2007 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:ifwedp:6166&r=ets |
| By: | McCauley, Joseph L.; Bassler, Kevin E.; Gunaratne, Gemunu h. |
| Abstract: | The condition for stationary increments, not scaling, detemines long time pair autocorrelations. An incorrect assumption of stationary increments generates spurious stylized facts, fat tails and a Hurst exponent Hs=1/2, when the increments are nonstationary, as they are in FX markets. The nonstationarity arises from systematic uneveness in noise traders’ behavior. Spurious results arise mathematically from using a log increment with a ‘sliding window’. We explain why a hard to beat market demands martingale dynamics , and martingales with nonlinear variance generate nonstationary increments. The nonstationarity is exhibited directly for Euro/Dollar FX data. We observe that the Hurst exponent Hs generated by the using the sliding window technique on a time series plays the same role as does Mandelbrot’s Joseph exponent. Finally, Mandelbrot originally assumed that the ‘badly behaved second moment of cotton returns is due to fat tails, but that nonconvergent behavior is instead direct evidence for nonstationary increments. Summarizing, the evidence for scaling and fat tails as the basis for econophysics and financial economics is provided neither by FX markets nor by cotton price data. |
| Keywords: | Nonstationary increments; martingales; fat tails; Hurst exponent scaling |
| JEL: | C40 G15 |
| Date: | 2007–10–12 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:5303&r=ets |