D. Allen, M. McAleer, R. Powell, Abhay Kumar-Singh
This paper presents an application of a recently developed approach by Matteson and James (2012) for the analysis of change points in a data set, namely major financial market indices converted to financial return series. The general problem concerns the inference of a change in the distribution of a set of time-ordered variables. The approach involves the nonparametric estimation of both the number of change points and the positions at which they occur. The approach is general and does not involve assumptions about the nature of the distributions involved or the type of change beyond the assumption of the existence of the absolute moment, for some 2 (0; 2). The estimation procedure is based on hierarchical clustering and the application of both divisive and agglomerative algorithms. The method is used to evaluate the impact of the Global Financial Crisis (GFC) on the US, French, German, UK, Japanese and Chinese markets, as represented by the S&P500, CAC, DAX, FTSE All Share, Nikkei 225 and Shanghai A share Indices, respectively, from 2003 to 2013. The approach is used to explore the timing and number of change points in the datasets corresponding to the GFC and subsequent European Debt Crisis.
{"title":"Nonparametric Multiple Change Point Analysis of the Global Financial Crisis","authors":"D. Allen, M. McAleer, R. Powell, Abhay Kumar-Singh","doi":"10.2139/ssrn.2270029","DOIUrl":"https://doi.org/10.2139/ssrn.2270029","url":null,"abstract":"This paper presents an application of a recently developed approach by Matteson and James (2012) for the analysis of change points in a data set, namely major financial market indices converted to financial return series. The general problem concerns the inference of a change in the distribution of a set of time-ordered variables. The approach involves the nonparametric estimation of both the number of change points and the positions at which they occur. The approach is general and does not involve assumptions about the nature of the distributions involved or the type of change beyond the assumption of the existence of the absolute moment, for some 2 (0; 2). The estimation procedure is based on hierarchical clustering and the application of both divisive and agglomerative algorithms. The method is used to evaluate the impact of the Global Financial Crisis (GFC) on the US, French, German, UK, Japanese and Chinese markets, as represented by the S&P500, CAC, DAX, FTSE All Share, Nikkei 225 and Shanghai A share Indices, respectively, from 2003 to 2013. The approach is used to explore the timing and number of change points in the datasets corresponding to the GFC and subsequent European Debt Crisis.","PeriodicalId":264857,"journal":{"name":"ERN: Semiparametric & Nonparametric Methods (Topic)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128559760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper shows how to compute the h-step-ahead predictive likelihood for any subset of the observed variables in parametric discrete time series models estimated with Bayesian methods. The subset of variables may vary across forecast horizons and the problem thereby covers marginal and joint predictive likelihoods for a fixed subset as special cases. The basic idea is to utilize well-known techniques for handling missing data when computing the likelihood function, such as a missing observations consistent Kalman filter for linear Gaussian models, but it also extends to nonlinear, nonnormal state-space models. The predictive likelihood can thereafter be calculated via Monte Carlo integration using draws from the posterior distribution. As an empirical illustration, we use euro area data and compare the forecasting performance of the New Area-Wide Model, a small-open-economy DSGE model, to DSGEVARs, and to reduced-form linear Gaussian models. JEL Classification: C11, C32, C52, C53, E37
{"title":"Predictive Likelihood Comparisons with DSGE and DSGE-VAR Models","authors":"A. Warne, G. Coenen, K. Christoffel","doi":"10.2139/ssrn.2250968","DOIUrl":"https://doi.org/10.2139/ssrn.2250968","url":null,"abstract":"This paper shows how to compute the h-step-ahead predictive likelihood for any subset of the observed variables in parametric discrete time series models estimated with Bayesian methods. The subset of variables may vary across forecast horizons and the problem thereby covers marginal and joint predictive likelihoods for a fixed subset as special cases. The basic idea is to utilize well-known techniques for handling missing data when computing the likelihood function, such as a missing observations consistent Kalman filter for linear Gaussian models, but it also extends to nonlinear, nonnormal state-space models. The predictive likelihood can thereafter be calculated via Monte Carlo integration using draws from the posterior distribution. As an empirical illustration, we use euro area data and compare the forecasting performance of the New Area-Wide Model, a small-open-economy DSGE model, to DSGEVARs, and to reduced-form linear Gaussian models. JEL Classification: C11, C32, C52, C53, E37","PeriodicalId":264857,"journal":{"name":"ERN: Semiparametric & Nonparametric Methods (Topic)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130376931","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A Bayesian semi-parametric estimation of the binary response model using Markov Chain Monte Carlo algorithms is proposed. The performances of the parametric and semi-parametric models are presented. The mean squared errors, receiver operating characteristic curve, and the marginal effect are used as the model selection criteria. Simulated data and Monte Carlo experiments show that unless the binary data is extremely unbalanced the semi-parametric and parametric models perform equally well. However, if the data is extremely unbalanced the maximum likelihood estimation does not converge whereas the Bayesian algorithms do. An application is also presented.
{"title":"Comparison of Bayesian and Sample Theory Semi-Parametric Binary Response Model","authors":"Xiangjin Shen, H. Tsurumi, Shiliang Li","doi":"10.2139/ssrn.2294625","DOIUrl":"https://doi.org/10.2139/ssrn.2294625","url":null,"abstract":"A Bayesian semi-parametric estimation of the binary response model using Markov Chain Monte Carlo algorithms is proposed. The performances of the parametric and semi-parametric models are presented. The mean squared errors, receiver operating characteristic curve, and the marginal effect are used as the model selection criteria. Simulated data and Monte Carlo experiments show that unless the binary data is extremely unbalanced the semi-parametric and parametric models perform equally well. However, if the data is extremely unbalanced the maximum likelihood estimation does not converge whereas the Bayesian algorithms do. An application is also presented.","PeriodicalId":264857,"journal":{"name":"ERN: Semiparametric & Nonparametric Methods (Topic)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2013-04-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122861905","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The method for simulation of labor market participation used in the LIMTIP models for Argentina, Chile, and Mexico is described. In each case, all eligible adults not working full-time were assigned full-time jobs. In all households that included job recipients, the time spent on household production was imputed for everyone included in the time-use survey. The feasibility of assessing the quality of the simulations is discussed. For each simulation, the recipient group is compared to the donor group, both in terms of demographic similarity and in terms of the imputed usual hours, earnings, and household production produced in the simulation. In each case, the simulations are of reasonable quality, given the nature of the challenges in assessing their quality.
{"title":"Simulations of Full-Time Employment and Household Work in the Levy Institute Measure of Time and Income Poverty (LIMTIP) for Argentina, Chile, and Mexico","authors":"Thomas Masterson","doi":"10.2139/ssrn.2101820","DOIUrl":"https://doi.org/10.2139/ssrn.2101820","url":null,"abstract":"The method for simulation of labor market participation used in the LIMTIP models for Argentina, Chile, and Mexico is described. In each case, all eligible adults not working full-time were assigned full-time jobs. In all households that included job recipients, the time spent on household production was imputed for everyone included in the time-use survey. The feasibility of assessing the quality of the simulations is discussed. For each simulation, the recipient group is compared to the donor group, both in terms of demographic similarity and in terms of the imputed usual hours, earnings, and household production produced in the simulation. In each case, the simulations are of reasonable quality, given the nature of the challenges in assessing their quality.","PeriodicalId":264857,"journal":{"name":"ERN: Semiparametric & Nonparametric Methods (Topic)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128080946","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Along the lines of Janssen's and Pfanzagl's work the testing theory for statistical functionals is further developed for non-parametric one-sample problems. Efficient tests for the one-sided and two-sided problems are derived for nonparametric statistical functionals. The asymptotic power function is calculated under implicit alternatives and hypotheses, which are given by the functional itself, for the one-sided and two-sided cases. Under mild regularity assumptions is shown that these tests are asymptotic most powerful. The combination of the modern theory of Le Cam and approximation in limit experiments provide a deep insight into the upper bounds for asymptotic power functions tests for the one-sided and two-sided problems of hypothesis testing. As example tests concerning the von Mises functional are treated in nonparametric context.
{"title":"Semiparametric Testing of Statistical Functionals Revisited","authors":"V. Ostrovski","doi":"10.2139/ssrn.2287633","DOIUrl":"https://doi.org/10.2139/ssrn.2287633","url":null,"abstract":"Abstract Along the lines of Janssen's and Pfanzagl's work the testing theory for statistical functionals is further developed for non-parametric one-sample problems. Efficient tests for the one-sided and two-sided problems are derived for nonparametric statistical functionals. The asymptotic power function is calculated under implicit alternatives and hypotheses, which are given by the functional itself, for the one-sided and two-sided cases. Under mild regularity assumptions is shown that these tests are asymptotic most powerful. The combination of the modern theory of Le Cam and approximation in limit experiments provide a deep insight into the upper bounds for asymptotic power functions tests for the one-sided and two-sided problems of hypothesis testing. As example tests concerning the von Mises functional are treated in nonparametric context.","PeriodicalId":264857,"journal":{"name":"ERN: Semiparametric & Nonparametric Methods (Topic)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2012-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131413879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We provide necessary and sufficient conditions for the local identification of the finite dimensional parameters in a semiparametric dynamic discrete choice model under additive separability and conditional independence assumption (Rust (1987)). We show that the policy value approach commonly used in the two-step estimation methodologies has convenient features so that the conditional version of Rothenberg's (1971) parametric identification results can be readily applied. We provide results for both the single agent problems and a class of games of incomplete information. These conditions are easy to check under the extreme value distributional assumption and when the payoff function has a linear-in-parameter specification. Our approach does not depend on the discreteness of the control variable and can be used to derive analogous conditions in other Markov decision models. Our approach can also be used when the value of the discounting factor not known.
{"title":"Local Identification in Markov Decision Models","authors":"Sorawoot Srisuma","doi":"10.2139/ssrn.1865152","DOIUrl":"https://doi.org/10.2139/ssrn.1865152","url":null,"abstract":"We provide necessary and sufficient conditions for the local identification of the finite dimensional parameters in a semiparametric dynamic discrete choice model under additive separability and conditional independence assumption (Rust (1987)). We show that the policy value approach commonly used in the two-step estimation methodologies has convenient features so that the conditional version of Rothenberg's (1971) parametric identification results can be readily applied. We provide results for both the single agent problems and a class of games of incomplete information. These conditions are easy to check under the extreme value distributional assumption and when the payoff function has a linear-in-parameter specification. Our approach does not depend on the discreteness of the control variable and can be used to derive analogous conditions in other Markov decision models. Our approach can also be used when the value of the discounting factor not known.","PeriodicalId":264857,"journal":{"name":"ERN: Semiparametric & Nonparametric Methods (Topic)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132363921","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
H. Koster, Jos N. van Ommeren, Piet Rietveld(Deceased)
This paper examines the effects of specialisation (within-sector clustering) and diversity (between-sector clustering) on business services profitability and location choice. We apply a semiparametric Poisson sorting model allowing for firm-specific effects. We find that for most firms, profitability of business services firms is substantially higher close to specialised clusters of business services firms. A standard deviation increase in business services specialisation leads to on average a 40 percent increase in the probability that a business services firm locates there, supporting theories of Marshall, Arrow and Romer. It is also profitable for most business services firms to locate near a group of firms that belong to the same sector, not necessarily business services firms, so diversity is negatively related to location decisions. Almost all firms either benefit from within-sector clustering or between-sector clustering. Within-sector clusters are particularly profitable for large mature firms, whereas between-sector clusters are relatively more profitable for smaller innovative firms.
{"title":"Geographic Concentration of Business Services Firms: A Poisson Sorting Model","authors":"H. Koster, Jos N. van Ommeren, Piet Rietveld(Deceased)","doi":"10.2139/ssrn.1853232","DOIUrl":"https://doi.org/10.2139/ssrn.1853232","url":null,"abstract":"This paper examines the effects of specialisation (within-sector clustering) and diversity (between-sector clustering) on business services profitability and location choice. We apply a semiparametric Poisson sorting model allowing for firm-specific effects. We find that for most firms, profitability of business services firms is substantially higher close to specialised clusters of business services firms. A standard deviation increase in business services specialisation leads to on average a 40 percent increase in the probability that a business services firm locates there, supporting theories of Marshall, Arrow and Romer. It is also profitable for most business services firms to locate near a group of firms that belong to the same sector, not necessarily business services firms, so diversity is negatively related to location decisions. Almost all firms either benefit from within-sector clustering or between-sector clustering. Within-sector clusters are particularly profitable for large mature firms, whereas between-sector clusters are relatively more profitable for smaller innovative firms.","PeriodicalId":264857,"journal":{"name":"ERN: Semiparametric & Nonparametric Methods (Topic)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126646006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We consider a difference based ridge regression estimator and a Liu type estimator of the regression parameters in the partial linear semiparametric regression model, [email protected][email protected] Both estimators are analyzed and compared in the sense of mean-squared error. We consider the case of independent errors with equal variance and give conditions under which the proposed estimators are superior to the unbiased difference based estimation technique. We extend the results to account for heteroscedasticity and autocovariance in the error terms. Finally, we illustrate the performance of these estimators with an application to the determinants of electricity consumption in Germany.
{"title":"Difference Based Ridge and Liu Type Estimators in Semiparametric Regression Models","authors":"E. Duran, W. Härdle, M. Osipenko","doi":"10.2139/ssrn.2894251","DOIUrl":"https://doi.org/10.2139/ssrn.2894251","url":null,"abstract":"We consider a difference based ridge regression estimator and a Liu type estimator of the regression parameters in the partial linear semiparametric regression model, [email protected][email protected] Both estimators are analyzed and compared in the sense of mean-squared error. We consider the case of independent errors with equal variance and give conditions under which the proposed estimators are superior to the unbiased difference based estimation technique. We extend the results to account for heteroscedasticity and autocovariance in the error terms. Finally, we illustrate the performance of these estimators with an application to the determinants of electricity consumption in Germany.","PeriodicalId":264857,"journal":{"name":"ERN: Semiparametric & Nonparametric Methods (Topic)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125937294","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Regression analyses of cross-country economic growth data are complicated by two main forms of model uncertainty: the uncertainty in selecting explanatory variables and the uncertainty in specifying the functional form of the regression function. Most discussions in the literature address these problems independently, yet a joint treatment is essential. We perform this joint treatment by extending the linear model to allow for multiple-regime parameter heterogeneity of the type suggested by new growth theory, while addressing the variable selection problem by means of Bayesian model averaging. Controlling for variable selection uncertainty, we confirm the evidence in favor of new growth theory presented in several earlier studies. However, controlling for functional form uncertainty, we find that the effects of many of the explanatory variables identified in the literature are not robust across countries and variable selections.
{"title":"Variable Selection and Functional Form Uncertainty in Cross-Country Growth Regressions","authors":"Tim Salimans","doi":"10.2139/ssrn.1742868","DOIUrl":"https://doi.org/10.2139/ssrn.1742868","url":null,"abstract":"Regression analyses of cross-country economic growth data are complicated by two main forms of model uncertainty: the uncertainty in selecting explanatory variables and the uncertainty in specifying the functional form of the regression function. Most discussions in the literature address these problems independently, yet a joint treatment is essential. We perform this joint treatment by extending the linear model to allow for multiple-regime parameter heterogeneity of the type suggested by new growth theory, while addressing the variable selection problem by means of Bayesian model averaging. Controlling for variable selection uncertainty, we confirm the evidence in favor of new growth theory presented in several earlier studies. However, controlling for functional form uncertainty, we find that the effects of many of the explanatory variables identified in the literature are not robust across countries and variable selections.","PeriodicalId":264857,"journal":{"name":"ERN: Semiparametric & Nonparametric Methods (Topic)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130698955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：