Essays on the econometrics of macroeconomic survey data
| Autor: | Conflitti, Cristina |
|---|---|
| Jazyk: | francouzština |
| Rok vydání: | 2012 |
| Předmět: | |
| Druh dokumentu: | Doctoral Thesis |
| Popis: | This thesis contains three essays covering different topics in the field of statisticsand econometrics of survey data. Chapters one and two analyse two aspectsof the Survey of Professional Forecasters (SPF hereafter) dataset. This surveyprovides a large information on macroeconomic expectations done by the professionalforecasters and offers an opportunity to exploit a rich information set.But it poses a challenge on how to extract the relevant information in a properway. The last chapter addresses the issue of analyzing the opinions on the euroreported in the Flash Eurobaromenter dataset.The first chapter Measuring Uncertainty and Disagreement in the EuropeanSurvey of Professional Forecasters proposes a density forecast methodology basedon the piecewise linear approximation of the individual’s forecasting histograms,to measure uncertainty and disagreement of the professional forecasters. Since1960 with the introduction of the SPF in the US, it has been clear that they were auseful source of information to address the issue on how to measure disagreementand uncertainty, without relying on macroeconomic or time series models. Directmeasures of uncertainty are seldom available, whereas many surveys report pointforecasts from a number of individual respondents. There has been a long traditionof using measures of the dispersion of individual respondents’ point forecasts(disagreement or consensus) as proxies for uncertainty. Unlike other surveys, theSPF represents an exception. It directly asks for the point forecast, and for theprobability distribution, in the form of histogram, associated with the macro variablesof interest. An important issue that should be considered concerns how toapproximate individual probability densities and get accurate individual resultsfor disagreement and uncertainty before computing the aggregate measures. Incontrast to Zarnowitz and Lambros (1987), and Giordani and Soderlind (2003) weovercome the problem associated with distributional assumptions of probabilitydensity forecasts by using a non parametric approach that, instead of assuminga functional form for the individual probability law, approximates the histogramby a piecewise linear function. In addition, and unlike earlier works that focus onUS data, we employ European data, considering gross domestic product (GDP),inflation and unemployment.The second chapter Optimal Combination of Survey Forecasts is based ona joint work with Christine De Mol and Domenico Giannone. It proposes anapproach to optimally combine survey forecasts, exploiting the whole covariancestructure among forecasters. There is a vast literature on forecast combinationmethods, advocating their usefulness both from the theoretical and empiricalpoints of view (see e.g. the recent review by Timmermann (2006)). Surprisingly,it appears that simple methods tend to outperform more sophisticated ones, asshown for example by Genre et al. (2010) on the combination of the forecasts inthe SPF conducted by the European Central Bank (ECB). The main conclusion ofseveral studies is that the simple equal-weighted average constitutes a benchmarkthat is hard to improve upon. In contrast to a great part of the literature whichdoes not exploit the correlation among forecasters, we take into account the fullcovariance structure and we determine the optimal weights for the combinationof point forecasts as the minimizers of the mean squared forecast error (MSFE),under the constraint that these weights are nonnegative and sum to one. Wecompare our combination scheme with other methodologies in terms of forecastingperformance. Results show that the proposed optimal combination scheme is anappropriate methodology to combine survey forecasts.The literature on point forecast combination has been widely developed, howeverthere are fewer studies analyzing the issue for combination density forecast.We extend our work considering the density forecasts combination. Moving fromthe main results presented in Hall and Mitchell (2007), we propose an iterativealgorithm for computing the density weights which maximize the average logarithmicscore over the sample period. The empirical application is made for theEuropean GDP and inflation forecasts. Results suggest that optimal weights,obtained via an iterative algorithm outperform the equal-weighted used by theECB density combinations.The third chapter entitled Opinion surveys on the euro: a multilevel multinomiallogistic analysis outlines the multilevel aspects related to public attitudestoward the euro. This work was motivated by the on-going debate whether theperception of the euro among European citizenships after ten years from its introductionwas positive or negative. The aim of this work is, therefore, to disentanglethe issue of public attitudes considering either individual socio-demographic characteristicsand macroeconomic features of each country, counting each of themas two separate levels in a single analysis. Considering a hierarchical structurerepresents an advantage as it models within-country as well as between-countryrelations using a single analysis. The multilevel analysis allows the considerationof the existence of dependence between individuals within countries induced byunobserved heterogeneity between countries, i.e. we include in the estimationspecific country characteristics not directly observable. In this chapter we empiricallyinvestigate which individual characteristics and country specificities aremost important and affect the perception of the euro. The attitudes toward theeuro vary across individuals and countries, and are driven by personal considerationsbased on the benefits and costs of using the single currency. Individualfeatures, such as a high level of education or living in a metropolitan area, havea positive impact on the perception of the euro. Moreover, the country-specificeconomic condition can influence individuals attitudes. Doctorat en Sciences économiques et de gestion info:eu-repo/semantics/nonPublished |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|