|
on Econometrics |
| By: | Hoonhout, P.J.M. |
| Abstract: | This thesis deals with attrition in panel data. The problem associated with attrition is that it can lead to estimation results that suffer from selection bias. This can be avoided by using attrition models that are sufficiently unrestrictive to allow for a wide range of potential selection. In chapter 2, I propose the Sequential Additively Nonignorable (SAN) attrition model. This model combines an Additive Nonignorability assumption with the Sequential Attrition assumption, to just-identify the joint population distribution in Panel data with any number of waves. The identification requires the availability of refreshment samples. Just-identification means that the SAN model has no testable implications. In other words, less restrictive identified models do not exist. To estimate SAN models, I propose a weighted Generalized Method of Moments estimator, and derive its repeated sampling behaviour in large samples. This estimator is applied to the Dutch Transportation Panel and the English Longitudinal Study of Ageing. In chapter 4, a likelihood-based alternative estimation approach is proposed, by means of an EM algorithm. Maximum Likelihood estimates can be useful if it is hard to obtain an explicit expression for the score function implied by the likelihood. In that case, the weighted GMM approach is not applicable. |
| Date: | 2011–06–28 |
| URL: | http://d.repec.org/n?u=RePEc:ner:ucllon:http://discovery.ucl.ac.uk/1318085/&r=ecm |
| By: | Tomás del Barrio Castro; Paulo M.M. Rodrigues; A. M. Robert Taylor |
| Abstract: | In this paper we investigate the impact of non-stationary cycles on the asymptotic and finite sample properties of standard unit root tests. Results are presented for the augmented Dickey-Fuller normalised bias and t-ratio-based tests (Dickey and Fuller, 1979, and Said and Dickey, 1984), the variance ratio unit root test of Breitung (2002) and the M class of unit-root tests introduced by Stock (1999) and Perron and Ng (1996). The limiting distributions of these statistics are derived in the presence of non-stationary cycles. We show that while the ADF statistics remain pivotal (provided the test regression is properly augmented), this is not the case for the other statistics considered and show numerically that the size properties of the tests based on these statistics are too unreliable to be used in practice. We also show that the t-ratios associated with lags of the dependent variable of order greater than two in the ADF regression are asymptotically normally distributed. This is an important result as it implies that extant sequential methods (see Hall, 1994 and Ng and Perron, 1995) used to determine the order of augmentation in the ADF regression remain valid in the presence of non-stationary cycles. |
| JEL: | C20 C22 |
| Date: | 2011 |
| URL: | http://d.repec.org/n?u=RePEc:ptu:wpaper:w201124&r=ecm |
| By: | Daniel Bartz; Kerr Hatrick; Christian W. Hesse; Klaus-Robert M\"uller; Steven Lemm |
| Abstract: | Robust and reliable covariance estimation plays a decisive role in financial applications. An important class of estimators is based on Factor models. Here, we show by extensive Monte Carlo simulations that covariance matrices derived from the statistical Factor Analysis model exhibit a systematic error, which is similar to the well-known systematic error of the spectrum of the sample covariance matrix. Moreover, we introduce the Directional Variance Adjustment (DVA) algorithm, which diminishes the systematic error. In a thorough empirical study for the US, European, and Hong Kong market we evince that our proposed method leads to improved portfolio allocation. |
| Date: | 2011–09 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1109.3069&r=ecm |
| By: | Raffaella Calabrese (Geary Institute, University College Dublin); Silvia Angela Osmetti (Department of Statistics, University Cattolica del Dacro Cuore, Milan) |
| Abstract: | The most used regression model with binary dependent variable is the logistic regression model. When the dependent variable represents a rare event, the logistic regression model shows relevant drawbacks. In order to overcome these drawbacks we propose the Generalized Extreme Value (GEV) regression model. In particular, in a Generalized Linear Model (GLM) with binary dependent variable we suggest the quantile function of the GEV distribution as link function, so our attention is focused on the tail of the response curve for values close to one. The estimation procedure is the maximum likelihood method. This model accommodates skewness and it presents a generalization of GLMs with log-log link function. In credit risk analysis a pivotal topic is the default probability estimation. Since defaults are rare events, we apply the GEV regression to empirical data on Italian Small and Medium Enterprises (SMEs) to model their default probabilities. |
| Date: | 2011–09–15 |
| URL: | http://d.repec.org/n?u=RePEc:ucd:wpaper:201120&r=ecm |
| By: | Biørn, Erik (Dept. of Economics, University of Oslo); R. Wangen, Knut (Department of Health Management and Health Economics, University of Oslo) |
| Abstract: | The aim of this paper is two-fold: (a) to establish a ‘map’ for describing the wide class of Limited Dependent Variables (LDV) univariate and multivariate models in the econometric literature and (b) to localize typical models in this tradition within the structure, extending typologies of Heckman (1976) and Amemiya (1984). The classification system, or language, proposed, is given at different level of detail. Its scope is (1) that the latent variables can have any parametric distribution, (2) that a set of observation rules which include the observed, censored, missing status, is imposed, (3) that it should be possible to write a model combining (1) and (2) by means of a computer algorithm, also potentially applicable for generating samples and (4) that the models belonging to the system should have names to facilitate communication among researchers. The likelihood functions corresponding to the models’ observed endogenous variables are discussed, but the paper is not concerned with describing the application of these functions for inference. |
| Keywords: | Micro-econometrics; Limited dependent variables; Latent variables; Discrete choice; Censoring; Truncation; Missing observations |
| JEL: | C16 C24 C25 C34 C35 C51 |
| Date: | 2011–09–14 |
| URL: | http://d.repec.org/n?u=RePEc:hhs:osloec:2011_018&r=ecm |
| By: | Maria Elvira Mancino (Dipartimento di Matematica per le Decisioni, Universita' degli Studi di Firenze); Simona Sanfelici (Dipartimento di Economia, Universita' di Parma) |
| Abstract: | We propose a new methodology based on Fourier analysis to estimate the fourth power of volatility function (spot quarticity) and, as a byproduct, the integrated function. We prove consistency of the proposed estimator of integrated quarticity. Further we analyze its efficiency in the presence of microstructure noise, both from a theoretical and empirical viewpoint. Extensions to higher powers of volatility and to the multivariate case are also discussed. |
| Keywords: | volatility, covariance, quarticity, microstructure, Fourier analysis |
| Date: | 2011–09 |
| URL: | http://d.repec.org/n?u=RePEc:flo:wpaper:2011-06&r=ecm |
| By: | Max Köhler (Georg-August-University Göttingen); Anja Schindler (Georg-August-University Göttingen); Stefan Sperlich (Université de Genéve) |
| Abstract: | Over the last four decades, several methods for selecting the smoothing parameter, generally called the bandwidth, have been introduced in kernel regression. They differ quite a bit, and although there already exist more selection methods than for any other regression smoother we can still see coming up new ones. Given the need of automatic data-driven bandwidth selectors for applied statistics, this review is intended to explain and compare these methods. |
| Keywords: | Kernel regression estimation; Bandwidth Selection; Plug-in; Cross Validation |
| Date: | 2011–09–13 |
| URL: | http://d.repec.org/n?u=RePEc:got:gotcrc:095&r=ecm |
| By: | Paul S. Clarke; Tom M. Palmer; Frank Windmeijer (Institute for Fiscal Studies and University of Bristol) |
| Abstract: | <p>Instrumental variables analysis using genetic markers as instruments is now a widely used technique in epidemiology and biostatistics. As single markers tend to explain only a small proportion of phenotypical variation, there is increasing interest in using multiple genetic markers to obtain more precise estimates of causal parameters. Structural mean models (SMMs) are semi-parametric models that use instrumental variables to identify causal parameters, but there has been little work on using these models with multiple instruments, particularly for multiplicative and logistic SMMs. In this paper, we show how additive, multiplicative and logistic SMMs with multiple discrete instrumental variables can be estimated efficiently using the generalised method of moments (GMM) estimator, how the Hansen J-test can be used to test for model mis-specification, and how standard GMM software routines can be used to fit SMMs. We further show that multiplicative SMMs, like the additive SMM, identify a weighted average of local causal effects if selection is monotonic. We use these methods to reanalyse a study of the relationship between adiposity and hypertension using SMMs with two genetic markers as instruments for adiposity. We find strong effects of adiposity on hypertension, but no evidence of unobserved confounding.</p> |
| Date: | 2011–08 |
| URL: | http://d.repec.org/n?u=RePEc:ifs:cemmap:28/11&r=ecm |
| By: | Steffen Andersen (Copenhagen Business School); Glenn W. Harrison (Robinson College of Business, Georgia State University); Morten Lau (Durham Business School); Elisabet E. Rutstroem (Robinson College of Business, Georgia State University) |
| Abstract: | We develop an extension of the familiar linear mixed logit model to allow for the direct estimation of parametric non-linear functions defined over structural parameters. Classic applications include the estimation of coefficients of utility functions to characterize risk attitudes and discounting functions to characterize impatience. There are several unexpected benefits of this extension, apart from the ability to directly estimate structural parameters of theoretical interest. |
| Date: | 2011–01–01 |
| URL: | http://d.repec.org/n?u=RePEc:dur:durham:2011_04&r=ecm |
| By: | Bai, Jushan; Wang, Peng |
| Abstract: | Motivated by the great moderation in major U.S. macroeconomic time series, we formulate the regime switching problem through a conditional Markov chain. We model the long-run volatility change as a recurrent structure change, while short-run changes in the mean growth rate as regime switches. Both structure and regime are unobserved. The structure is assumed to be Markovian. Conditioning on the structure, the regime is also Markovian, whose transition matrix is structure-dependent. This formulation imposes interpretable restrictions on the Hamilton Markov switching model. Empirical studies show that this restricted model well identifies both short-run regime switches and long-run structure changes in the U.S. macroeconomic data. |
| Keywords: | Markov regime switching; Conditional Markov chain |
| JEL: | C32 C22 C01 |
| Date: | 2011–08 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:33369&r=ecm |
| By: | Tomás del Barrio Castro; Denise R. Osborn; A.M. Robert Taylor |
| Date: | 2011 |
| URL: | http://d.repec.org/n?u=RePEc:man:sespap:1121&r=ecm |