|
on Econometrics |
| By: | Isabel Molina; Nicola Salvati; Monica Pratesi |
| Abstract: | This work assumes that the small area quantities of interest follow a Fay-Herriot model with spatially correlated random area effects. Under this model, parametric and nonparametric bootstrap procedures are proposed for estimating the mean squared error of the EBLUP (Empirical Best Linear Unbiased Predictor). A simulation study compares the bootstrap estimates with an asymptotic analytical approximation and studies the robustness to non-normality. Finally, two applications with real data are described. |
| Date: | 2007–04 |
| URL: | http://d.repec.org/n?u=RePEc:cte:wsrepe:ws073408&r=ecm |
| By: | Miguel A. Segoviano Basurto |
| Abstract: | Portfolio credit risk measurement is greatly affected by data constraints, especially when focusing on loans given to unlisted firms. Standard methodologies adopt convenient, but not necessarily properly specified parametric distributions or simply ignore the effects of macroeconomic shocks on credit risk. Aiming to improve the measurement of portfolio credit risk, we propose the joint implementation of two new methodologies, namely the conditional probability of default (CoPoD) methodology and the consistent information multivariate density optimizing (CIMDO) methodology. CoPoD incorporates the effects of macroeconomic shocks into credit risk, recovering robust estimators when only short time series of loans exist. CIMDO recovers portfolio multivariate distributions (on which portfolio credit risk measurement relies) with improved specifications, when only partial information about borrowers is available. Implementation is straightforward and can be very useful in stress testing exercises (STEs), as illustrated by the STE carried out within the Danish Financial Sector Assessment Program. |
| Keywords: | Portfolio credit risk measurement , stress testing , macroeconomic shock measurement , multivariate density estimation , entropy distribution , Credit risk , Economic conditions , Statistics , Economic models , |
| Date: | 2007–01–03 |
| URL: | http://d.repec.org/n?u=RePEc:imf:imfwpa:06/283&r=ecm |
| By: | Man Yee Kan (Department of Sociology, University of Oxford); Stephen Pudney (Institute for Social and Economic Research) |
| Abstract: | We investigate the nature of measurement error in time use data. Analysis of 'stylised' recall questionnaire estimates and diary-based estimates of housework time from the same respondents gives evidence of systematic biases in the stylised estimates and large random errors in both types of data. We examine the effect of these measurement problems on three common types of statistical analyses in which the time use variable is used as: (i) a dependent variable, (ii) an explanatory variable, and (iii) a basis for cross-tabulations. We develop methods to correct the biases induced by these measurement errors. |
| Keywords: | housework, measurement error, time use |
| Date: | 2007–02 |
| URL: | http://d.repec.org/n?u=RePEc:ese:iserwp:2007-03&r=ecm |
| By: | Spierdijk, Laura (Dept. of Econometrics, University of Groningen); Voorneveld, Mark (Dept. of Economic Statistics, Stockholm School of Economics) |
| Abstract: | We provide three points of critique to the empirical analysis of Chung and Cox (1994, this Review) on the superstar phenomenon. They use a stochastic model which indicates that superstars may exist regardless of talent and that gives rise to the Yule distibution. Using a parametric bootstrap and several powerful test statistics, we find overwhelming evidence against the Yule distribution. The Yule distribution seems a fairly accurate approximation of the lower quantiles of the empirical distribution, but puts too much weight in the right tail of the distribution: it can model stardom, but fails to model superstardom. |
| Keywords: | Superstardom; Yule distribution |
| JEL: | J31 L82 Z11 |
| Date: | 2007–04–25 |
| URL: | http://d.repec.org/n?u=RePEc:hhs:hastef:0658&r=ecm |
| By: | Heather Laurie (Institute for Social and Economic Research) |
| Abstract: | This paper reports the results of an incentives experiment conducted at wave 14 of the British Household Panel Survey. A split-sample design was used to assess the effec of increasing the monetary incentive given to respondents on response rates. To be published as an ISER Working paper. |
| Keywords: | incentives, survey non-response |
| Date: | 2007–02 |
| URL: | http://d.repec.org/n?u=RePEc:ese:iserwp:2007-05&r=ecm |
| By: | Stefano Iacus (Department of Economics, Business and Statistics, University of Milan, IT); Nakahiro Yoshida (Graduate School of Mathematical Sciences, University of Tokyo) |
| Abstract: | The telegraph process {X(t), t>0}, is supposed to be observed at n+1 equidistant time points t_i=i Delta_n,i=0,1,... , n. The unknown value of lambda, the underlying rate of the Poisson process, is a parameter to be estimated. The asymptotic framework considered is the following: Delta_n -> 0, n Delta_n = T -> infty as n -> infty. We show that previously proposed moment type estimators are consistent and asymptotically normal but not efficient. We study further an approximated moment type estimator which is still not efficient but comes in explicit form. For this estimator the additional assumption n Delta_n^3 -> 0 is required in order to obtain asymptotic normality. Finally, we propose a new estimator which is consistent, asymptotically normal and asymptotically efficient under no additional hypotheses. |
| Keywords: | telegraph process, discretely observed process, inference for stochastic processes, |
| Date: | 2006–12–27 |
| URL: | http://d.repec.org/n?u=RePEc:bep:unimip:1045&r=ecm |
| By: | Giovanna Nicolini (Department of Economics, Business and Statistics); Francesca De Battisti (Department of Economics, Business and Statistics) |
| Abstract: | The application of the Rasch model to measure the Quality and Customer Satisfaction of a service is possible only for a single dimension of the service. But we know that the Quality and Customer Satisfaction of a service are a compound of more than one dimension. In such a case the Rasch method cannot be applied. As it is well known, the one dimension Rasch model supplies two graduated lists: the first is about subject satisfaction and the second concerns item quality. If there are K service dimensions and if the Rasch model is applied to each of them, K different graduated lists of subject satisfaction for the n individuals are obtained. The aim of this work is to define an overall individual satisfaction measure obtained as linear combination of the K subject graduated lists, with weights estimated in two different ways, for different hypothesis. We have shown that these two methods are consistent. |
| Keywords: | Rasch Model, Overall Satisfaction Measure, |
| Date: | 2006–10–19 |
| URL: | http://d.repec.org/n?u=RePEc:bep:unimip:1039&r=ecm |
| By: | Alessandro De Gregorio (Università di Milano, Italy) |
| Abstract: | We deal with a planar random flight {(X (t), Y (t)), 0 < t ? T } observed at n + 1 equidistant times ti = i?n , i = 0, 1, ..., n. The aim of this paper is to estimate the unknown value of the parameter ?, the underlying rate of the Poisson process. The planar random flights are not markovian, then we use an alternative argument to derive a pseudo-maximum likelihood estimator ? of the parameter ?. We consider two different types of asymptotic schemes and show the consistency, the asymptotic normality and efficiency of the estimator proposed. A Monte Carlo analysis for small sample size n permits us to analyze the empirical performance of ?. A different approach permits us to introduce an alternative estima- tor of ? which is consistent, asymptotically normal and asymptotically efficient without the request of other assumptions. |
| Keywords: | asymptotic efficiency, discretely observed process, planar random flight, inference for stochastic process, |
| Date: | 2007–03–19 |
| URL: | http://d.repec.org/n?u=RePEc:bep:unimip:1052&r=ecm |