|
on Econometrics |
| By: | Stephen P. Jenkins (Institute for Social and Economic Research); Markus Jäntti (University of Tampere) |
| Abstract: | This paper reviews methods for summarizing and comparing wealth distributions. We show that many of the tools commonly used to summarize income distributions can also be applied to wealth distributions, albeit adapted in order to account for the distinctive features of wealth distributions: zero and negative wealth values; spikes in density at or around zero; right-skewness with long and sparse tails combined with non-trivial prevalence of extreme values. Illustrations are provided using data for Finland. |
| Date: | 2005–05 |
| URL: | http://d.repec.org/n?u=RePEc:ese:iserwp:2005-05&r=ecm |
| By: | Annette Jäckle (Institute for Social and Economic Research) |
| Abstract: | Dependent interviewing techniques, where substantive information from previous interviews is fed forward and used in the formulation of questions or to prompt post-response edit checks, are increasingly employed by panel surveys. While there is substantial evidence that dependent interviewing improves the quality of longitudinal data, claims of improved efficiency of data collection and reduced respondent burden are mostly anecdotal. This paper uses data from a large experiment to systematically compare the effects of different question designs on efficiency and burden. The comparison highlights the wide variety of design options for dependent interviewing questions and their corresponding effects. In the present setup, efficiency gains were mainly due to reductions in coding costs for occupation and industry questions. The paper concludes by identifying the conditions under which dependent interviewing offers the largest scope for efficiency gains and burden reduction. |
| Keywords: | data collection, dependent interviewing |
| Date: | 2005–06 |
| URL: | http://d.repec.org/n?u=RePEc:ese:iserwp:2005-11&r=ecm |
| By: | Edvinsson, Rodney (Stockholm University) |
| Abstract: | For the period 1800 onwards, annual figures over GDP and GDP per capita for Sweden have been presented in different studies. For the 18th century no such annual series exist. The aim of this paper is to present annual data on GDP and GDP per capita in volume values for Sweden for the whole period 1720-1800. Only very rough estimates are provided, which are not based on any disaggregation of the different components of GDP. To estimate annual fluctuations, four different indicators are used: changes in the official accounts of harvests, marriage rates, the price of rye and import of un-milled grains. When investigating the long-term trends, the conclusion is that there was only a very modest increase in GDP per capita over the studied period. The growth of the GDP per capita became substantial not until the mid 19th century. However, GDP grew significantly during the studied period, but this growth mainly took the form of population growth. This in itself constituted a kind of technological progress, allowing a larger population per unit of land without a significant decrease in per capita production. |
| Keywords: | GDP; 18th century; Sweden; economic growth; national accounts; economic history |
| JEL: | E23 E32 J10 N13 N53 O10 O13 O47 O52 Q13 |
| Date: | 2005–06–14 |
| URL: | http://d.repec.org/n?u=RePEc:hhs:ratioi:0070&r=ecm |
| By: | D. S. Poskitt |
| Abstract: | Autoregressive models are commonly employed to analyze empirical time series. In practice, however, any autoregressive model will only be an approximation to reality and in order to achieve a reasonable approximation and allow for full generality the order of the autoregression, h say, must be allowed to go to infinity with T, the sample size. Although results are available on the estimation of autoregressive models when h increases indefinitely with T such results are usually predicated on assumptions that exclude (i) non-invertible processes and (ii) fractionally integrated processes. In this paper we will investigate the consequences of fitting long autoregressions under regularity conditions that allow for these two situations and where an infinite autoregressive representation of the process need not exist. Uniform convergence rates for the sample autocovariances are derived and corresponding convergence rates for the estimates of AR(h) approximations are established. A central limit theorem for the coefficient estimates is also obtained. An extension of a result on the predictive optimality of AIC to fractional and non-invertible processes is obtained. |
| Keywords: | Autoregression, Autoregressive approximation, Fractional process, Non-invertibility, Order selection, Asymptotic efficiency. |
| JEL: | C14 C32 C53 |
| Date: | 2005–06 |
| URL: | http://d.repec.org/n?u=RePEc:msh:ebswps:2005-16&r=ecm |
| By: | Farshid Jamshidian (Univ. of Twente) |
| Abstract: | This paper extends a recent martingale representation result of [N-S] for a L\'{e}vy process to filtrations generated by a rather large class of semimartingales. As in [N-S], we assume the underlying processes have moments of all orders, but here we allow angle brackets to be stochastic. Following their approach, including a chaotic expansion, and incorporating an idea of strong orthogonalization from [D], we show that the stable subspace generated by Teugels martingales is dense in the space of square-integrable martingales, yielding the representation. While discontinuities are of primary interest here, the special case of a (possibly infinite-dimensional) Brownian filtration is an easy consequence. |
| Keywords: | Martingale Representation, chaotic expansion, power brackets, Teugels martingales, Hilbert space, strong orthogonalization |
| JEL: | G |
| Date: | 2005–06–14 |
| URL: | http://d.repec.org/n?u=RePEc:wpa:wuwpfi:0506008&r=ecm |
| By: | Peter Lynn (Institute for Social and Economic Research); Emanuela Sala (Institute for Social and Economic Research) |
| Abstract: | Surveys that take repeat measurements on the same individuals (panel or follow-up surveys) are often used to measure change in employment characteristics. This article is concerned with measurement error in such estimates of change and, specifically, how the error might be reduced by the use of dependent interviewing (DI) techniques. We use data from a large-scale experiment that involved two interviews at an interval of around 17 months and compare estimates of change that are obtained using three different interviewing techniques: traditional independent interviewing (INDI), proactive dependent interviewing (PDI) and reactive dependent interviewing (RDI). We examine three characteristics of the respondent’s employment (occupation, employed status, and whether or not the respondent has managerial or supervisory responsibilities) and three characteristics of the employing organisation (industry, type of organisation, number of employees). We focus on the estimation of change in each of these six characteristics. We find that PDI results in lower levels of observed change for occupation, industry and number of employees. This reduction in observed change appears to represent a reduction in measurement error as the effect of PDI is particularly pronounced amongst respondents who have not reported a change in job between survey waves. Levels of change in employment characteristics amongst INDI respondents who have not reported a change in job remain implausibly high. The reduction in measurement error brought about by PDI is particularly associated with certain employment characteristics. A reduction in the observed level of change in occupation is associated with SOC major groups 1-4 and respondents working at workplaces with large number of employees. A reduction in the observed level of change in industry is associated with certain industries and with respondents who are managers or professionals (SOC major groups 1 or 2) or have foreman or supervisor status. A reduction with PDI in the observed level of change in number of employees at the workplace is associated with large workplaces, having foreman/supervisor status, being employed in the public administration or education sectors, and being in a craft or related occupation or a plant or machine operative. We also found that measurement error was particularly reduced by PDI amongst respondents aged 36 or over and amongst the most highly qualified respondents. |
| Keywords: | Measurement error, employment, interviewers, survey errors, survey methodology |
| Date: | 2004–12 |
| URL: | http://d.repec.org/n?u=RePEc:ese:iserwp:2004-26&r=ecm |
| By: | Stephen P. Jenkins (Institute for Social and Economic Research); Annette Jäckle (Institute for Social and Economic Research); Peter Lynn (Institute for Social and Economic Research); Emanuela Sala (Institute for Social and Economic Research) |
| Abstract: | This article is concerned with measurement error in panel survey reports of social security benefit receipt. Our aims are two-fold. First, we attempt to quantify the extent of measurement error and to identify its correlates. Second, we assess the extent to which this varies according to the questioning method used. Specifically, dependent interviewing has been proposed as a way to reduce under-reporting in some circumstances and we compare two versions of dependent interviewing (DI) with traditional independent interviewing in an experimental design. We use data from a large-scale UK household panel survey and we consider six benefits. To assess the measurement error, a validation exercise was conducted, with administrative data on benefit receipt matched at the individual level to the survey micro data. |
| Keywords: | Measurement error, dependent interviewing, longitudinal data quality, panel data |
| Date: | 2004–12 |
| URL: | http://d.repec.org/n?u=RePEc:ese:iserwp:2004-28&r=ecm |