|
on Econometrics |
| By: | Javier Alvarez; Manuel Arellano (CEMFI, Centro de Estudios Monetarios y Financieros) |
| Abstract: | We develop likelihood-based estimators for autoregressive panel data models that are consistent in the presence of time series heteroskedasticity. Bias corrected conditional score estimators, random effects maximum likelihood (RML) in levels and first differences, and estimators that impose mean stationarity and considered for AR(p) models with individual effects. We investigate identification under unit roots, and show that RML in levels may achieve substantial efficiency gains relative to estimators from data in differences. In an empirical application, we find evidence against unit roots in individual earnings processes from the PSID and the Spanish section of the European Panel. |
| Keywords: | Autoregressive panel data model, bias corrected score, time series heteroskedasticity, random effects, unit root identification, mean stationarity, individual earnings. |
| JEL: | C23 |
| Date: | 2004–12 |
| URL: | http://d.repec.org/n?u=RePEc:cmf:wpaper:wp2004_0421&r=ecm |
| By: | Christian Bauer |
| Abstract: | In this paper we consider the theoretical and empirical relevance of a new family of conditionally heteroskedastic models with a trend dependent conditional variance equation: the Trend-GARCH model. The interest in these models lies in the fact that modern microeco- nomic theory often suggests the connection between the past behavior of time series and the subsequent reaction of market individuals and thereon changes in the future characteristics of the time series. Our results reveal important properties of these models, which are con- sistent with stylized facts in ?financial data sets. They can also be employed for model identifi?cation, estimation, and testing. The em- pirical analysis of a broad variety of asset prices signi?ficantly supports the existence of trend effects. The Trend-GARCH model proves to be superior to alternative models such as EGARCH, AGARCH, or TGARCH in replicating the leverage effect in the conditional variance and in fi?tting the news impact curve. |
| Keywords: | GARCH, trend, volatility, news impact curve |
| JEL: | C22 C52 G12 |
| Date: | 2005–02 |
| URL: | http://d.repec.org/n?u=RePEc:uba:hadfwe:trend-garch-bauer_2005-02&r=ecm |
| By: | Zuleyca Díaz Martínez (Universidad Complutense de Madrid. Facultad de Económicas y Empresariales.Departamento de Economía Financiera y Contabilidad I.); José Fernández Menéndez (Universidad Complutense de Madrid. Facultad de Económicas y Empresariales.Departamento de Organización de Empresas.); Paloma Martínez Almodovar (Universidad Complutense de Madrid. Facultad de Económicas y Empresariales.Departamento de Organización de Empresas.) |
| Abstract: | Prediction of insurance companies insolvency has arised as an important problem in the field of financial research, due to the necessity of protecting the general public whilst minimizing the costs associated to this problem. Most methods applied in the past to tackle this question are traditional statistical techniques which use financial ratios as explicative variables. However, these variables do not usually satisfy statistical assumptions, what complicates the application of the mentioned methods.In this paper, a comparative study of the performance of a well-known parametric statistical technique (Linear Discriminant Analysis) and a non-parametric machine learning technique (See5) is carried out. We have applied the two methods to the problem of the prediction of insolvency of Spanish non-life insurance companies upon the basis of a set of financial ratios. Results indicate a higher performance of the machine learning technique, what shows that this method can be a useful tool to evaluate insolvency of insurance firms. |
| Date: | 2004 |
| URL: | http://d.repec.org/n?u=RePEc:ucm:doctra:04-12&r=ecm |
| By: | Mototsugu Shintani (Department of Economics, Vanderbilt University) |
| Abstract: | It has been claimed that the deviations from purchasing power parity are highly persistent and have quite long half-lives under the assumption of a linear adjustment of real exchange rates. However, inspired by trade cost models, nonlinear adjustment has been widely employed in recent empirical studies. This paper proposes a simple nonparametric procedure for evaluating the speed of adjustment in the presence of nonlinearity, using the largest Lyapunov exponent of the time series. The empirical result suggests that the speed of convergence to a long-run price level is indeed faster than what was found in previous studies with linear restrictions. |
| Keywords: | Mean reversion, nonlinear time series, nonparametric regression, purchasing power parity puzzle; Real exchange rates |
| JEL: | C14 C22 F31 |
| Date: | 2002–08 |
| URL: | http://d.repec.org/n?u=RePEc:van:wpaper:0219&r=ecm |
| By: | Xiaohong Chen (Department of Economics, New York University); Yanqin Fan (Department of Economics, Vanderbilt University) |
| Abstract: | This paper studies the estimation of a class of copula-based semiparametric stationary Markov models. These models are characterized by nonparametric invariant (or marginal) distributions and parametric copula functions that capture the temporal dependence of the processes; the implied transition distributions are all semiparametric, and a member in this class can be expressed as a generalized semiparametric regression transformation model. One advantage of this copula approach is to separate out the temporal dependence (such as clustering, tail dependence) from the marginal behavior (such as asymmetry, fat tails) of a time series. We present conditions under which processes generated by models in this class are beta-mixing; naturally, these conditions depend only on the copula specification. Simple estimators of the marginal distribution and the copula parameter are provided, and their asymptotic properties are established under easily verifiable conditions. These results allow us to easily obtain the root-n consistent and asymptotically normal estimators of important features of the transition distribution such as the (nonlinear) conditional moments and conditional quantiles. In addition, the semiparametric conditional quantile estimators are automatically monotonic across quantiles, which is attractive for portfolio conditional value-at-risk calculation. |
| Keywords: | Copula, beta-mixing, semiparametric estimation, parametric bootstrap |
| JEL: | C14 C22 |
| Date: | 2002–10 |
| URL: | http://d.repec.org/n?u=RePEc:van:wpaper:0226&r=ecm |