http://hdl.handle.net/10993/71 Search this channel search http://orbilu.uni.lu/simple-search http://hdl.handle.net/10993/40792 Title: On solutions to estimating equations and the empirical saddlepoint approximation of their distribution <br/> <br/>Author, co-author: Holcblat, Benjamin; Sowell, Fallaw Mon, 28 Oct 2019 14:27:32 GMT http://hdl.handle.net/10993/40785 Title: The Empirical Saddlepoint Estimator <br/> <br/>Author, co-author: Holcblat, Benjamin; Sowell, Fallaw Mon, 28 Oct 2019 14:11:25 GMT http://hdl.handle.net/10993/40784 Title: The Empirical Saddlepoint Estimator <br/> <br/>Author, co-author: Holcblat, Benjamin; Sowell, Fallaw Mon, 28 Oct 2019 14:11:03 GMT http://hdl.handle.net/10993/39918 Title: A nonparametric ACD model <br/> <br/>Author, co-author: Cosma, Antonio; Galli, Fausto <br/> <br/>Abstract: We propose a fully nonparametric approach to the analysis of the Autocorrelated Conditional Duration (ACD) process applied to durations between financial events. We use a recursive algorithm to estimate the nonparametric specification. In a Monte Carlo experiment, we analyse its forecasting performance and compare it with a correctly and a mis-specified parametric estimator. On a real dataset, the nonparametric estimator seems to mildly overperform in terms of predictive power. The nonparametric analysis can also provide guidance on the choice between alternative parametric specifications. In particular, once intraday seasonality is directly modelled in the conditional duration function, the nonparametric approach provides insights into the time-varying nature of the dynamics in the model that the standard procedures of deseasonalization may lead one to overlook. Tue, 16 Jul 2019 13:06:56 GMT http://hdl.handle.net/10993/39355 Title: Inference in Conditional Moment Restriction Models When there is Selection Due to Stratification <br/> <br/>Author, co-author: Cosma, Antonio; Kostyrka, André; Tripathi, Gautam <br/> <br/>Abstract: We show how to use a smoothed empirical likelihood approach to conduct efficient semiparametric inference in models characterized as conditional moment equalities when data is collected by variable probability sampling. Results from a simulation experiment suggest that the smoothed empirical likelihood based estimator can estimate the model parameters very well in small to moderately sized stratified samples. Mon, 15 Apr 2019 15:33:00 GMT http://hdl.handle.net/10993/37030 Title: Foreign Workers and the Wage Distribution: What Does the Influence Function Reveal? <br/> <br/>Author, co-author: Choe, Chung; Van Kerm, Philippe <br/> <br/>Abstract: This paper draws upon influence function regression methods to determine where foreign workers stand in the distribution of private sector wages in Luxembourg, and assess whether and how much their wages contribute to wage inequality. This is quantified by measuring the effect that a marginal increase in the proportion of foreign workers—foreign residents or cross-border workers—would have on selected quantiles and measures of inequality. Analysis of the 2006 Structure of Earnings Survey reveals that foreign workers have generally lower wages than natives and therefore tend to haul the overall wage distribution downwards. Yet, their influence on wage inequality reveals small and negative. All impacts are further muted when accounting for human capital and, especially, job characteristics. Not observing any large positive inequality contribution on the Luxembourg labour market is a striking result given the sheer size of the foreign workforce and its polarization at both ends of the skill distribution. Fri, 26 Oct 2018 03:30:02 GMT http://hdl.handle.net/10993/36954 Title: Economies of scale in recoverable robust maintenance location routing for rolling stock <br/> <br/>Author, co-author: Arts, Joachim; Tönissen, Denise <br/> <br/>Abstract: We consider the problem of locating maintenance facilities in a railway setting. Different facility sizes can be chosen for each candidate location and for each size there is an associated annual facility costs that can capture economies of scale in facility size. Because of the strategic nature of facility location, the opened facilities should be able to handle the current maintenance demand, but also the demand for any of the scenarios that can occur in the future. These scenarios capture changes such as changes to the line plan and the introduction of new rolling stock types. We allow recovery in the form of opening additional facilities, closing facilities, and increasing the facility size for each scenario. We provide a two-stage robust programming formulation. In the first-stage, we decide where to open what size of facility. In the second-stage, we solve a NP-hard maintenance location routing problem. We reformulate the problem as a mixed integer program that can be used to make an efficient column-and-constraint generation algorithm. To show that our algorithm works on practical sized instances, and to gain managerial insights, we perform a case study with instances from the Netherlands Railways. A counter intuitive insight is that economies of scale only play a limited role and that it is more important to reduce the transportation cost by building many small facilities, rather than a few large ones to profit from economies of scale. Mon, 15 Oct 2018 03:30:01 GMT http://hdl.handle.net/10993/36600 Title: Maintenance location routing for rolling stock under line and fleet planning uncertainty <br/> <br/>Author, co-author: Tönissen, Denise; Arts, Joachim; Shen, Zuo-Jun Mon, 17 Sep 2018 19:12:56 GMT http://hdl.handle.net/10993/36522 Title: Integrated likelihood based inference for nonlinear panel data models with unobserved effects <br/> <br/>Author, co-author: Schumann, Martin; Severini, Thomas A.; Tripathi, Gautam Tue, 04 Sep 2018 11:17:45 GMT http://hdl.handle.net/10993/36521 Title: Convexity of probit weights <br/> <br/>Author, co-author: Schumann, Martin; Tripathi, Gautam Tue, 04 Sep 2018 11:03:04 GMT http://hdl.handle.net/10993/36087 Title: Optimizing Trading Decisions for Hydro Storage Systems using Approximate Dual Dynamic Programming <br/> <br/>Author, co-author: Löhndorf, Nils; Wozabal, David; Minner, Stefan Thu, 05 Jul 2018 03:30:01 GMT http://hdl.handle.net/10993/35748 Title: Employment Dynamics, Firm Performance and Innovation Persistence in the Context of Differentiated Innovation Types: Evidence from Luxembourg <br/> <br/>Author, co-author: Zhen, Ni <br/> <br/>Abstract: This doctoral dissertation examines the essential topics of employment dynamics, firm performance and innovation persistence comprehensively. In particular, this doctoral dissertation provides an assessment of the differentiated role of innovation strategies in employment, firm performance and innovation persistence. Chapter 2 studies the dynamic relationship between technological innovation and employment using Luxembourgish firm level data pertaining to the non-financial corporate sector during the period 2003-2012. A simple theoretical model that distinguishes the employment effect of product innovation from that of process innovation is developed. The model is then estimated by two-step generalised method of moments using an unbalanced panel data stemming from the annual structural business survey merged with the biennial innovation survey. The results show that the semi-elasticity of employment with respect to product innovation lies between 0.2% and 0.5%. The differential in the employment effects between radical and incremental innovation is estimated to be 50%. Similarly, the employment level differential between product innovators and firms with unchanged products lies between 4% and 11%. Unlike product innovation, however, process innovation does not have any significant effect on the firm level of employment. Chapter 3 investigates the two-way relationship between technological innovation and firm performance at the firm level. In the framework of evolutionary economics, innovation is regarded as a highly cumulative process which exhibits positive feedback. This chapter aims at capturing the interdependent relationship and possible bidirectional causality between innovation and firm performance. Superior firm performance facilitates the emergence of innovations, innovation contributes to firm performance by gaining successful and sustainable competitive advantage, which forms a virtuous circle. A fully recursive simultaneous model is established where product and process innovation are explicitly distinguished. The system of simultaneous equations with mixed structure is estimated by full information maximum likelihood methods. The longitudinal firm-level data is applied over the 2003-2012 period by merging five waves of the Luxembourgish innovation survey with structural business surveys. This chapter discovers that enhanced firm performance facilitates process innovation and process innovation improves firm performance, which forms a self-reinforcing virtuous circle. An opposite pattern is identified for the product innovation on the ground of cannibalization effect and inherent market risks associated with new products. Chapter 4 explores innovation persistence at the firm level by means of dynamic nonlinear random effects models based on the estimator proposed by Albarrán et al. [2015]. It aims at capturing the true state dependence which indicates the causal relationship between innovation in one period and decision to innovate in the subsequent period. The Albarrán et al. [2015] method accounts for unobserved individual effects that are correlated with the initial conditions as well as the unbalanced structure of panel. Using five questionnaire waves of Luxembourgish Community Innovation Surveys (CIS) for the years 2002-2012, this study provides new insights on the differentiated patterns of persistence among product and process innovation. Results highlight the relevance of innovation persistence for all types of innovation, particularly the highest level of persistence is found for product innovation. In addition, the state dependence of product innovation is mainly associated with sunk costs relevant to R&D expenditures, whereas the state dependence of process innovation can be attributed to other factors such as dynamic increasing returns and learning effect. The further differentiation of product innovator category reveals that the state dependence of incremental product innovation can be mainly attributed to sunk costs relevant to R&D expenditures. In contrast, the joint significance of average R&D intensity, intramural R&D share as well as the past realization of radical product innovation suggests the role of other factors such as dynamic increasing returns and learning effect in fostering state dependence for radical innovations. Mon, 28 May 2018 15:01:46 GMT http://hdl.handle.net/10993/35092 Title: On Partial-Sum Processes of ARMAX Residuals <br/> <br/>Author, co-author: Holcblat, Benjamin; Gronneberg, Steffen <br/> <br/>Abstract: General and versatile results are established regarding the limit behavior of the partial-sum process of ARMAX residuals. Illustrations include ARMA with seasonal dummies, misspecified ARMAX models with autocorrelated errors, nonlinear ARMAX models, ARMA with a structural break, a wide range of ARMAX models with infinite-variance errors, weak GARCH models and the consistency of kernel estimation of the density of ARMAX errors. Our results identify the limit distributions, and provide a general algorithm to obtain pivot statistics for CUSUM tests. Thu, 01 Mar 2018 13:02:51 GMT http://hdl.handle.net/10993/35091 Title: The Empirical Saddlepoint Estimator <br/> <br/>Author, co-author: Holcblat, Benjamin; Sowell, Fallaw <br/> <br/>Abstract: Previous studies have shown that existing moment-based estimation approaches have poor small-sample performance in some applications. We propose an alternative that is based on the ESP (empirical saddlepoint) approximation of the solutions to the empirical moment conditions. Saddlepoint approximations are known to perform well in small sample. The novel estimator proposed, which we call the ESP estimator, is the mode of the ESP approximation. We show that it is consistent and asymptotically normal, and we study its higher-order bias. We propose novel test statistics based on the ESP estimator. Finally, we also investigate the finite-sample properties of the ESP estimator and related test statistics through Monte-Carlo simulations. Thu, 01 Mar 2018 13:02:36 GMT http://hdl.handle.net/10993/35086 Title: Income mobility profiles <br/> <br/>Author, co-author: Van Kerm, Philippe <br/> <br/>Abstract: An `income mobility profile' is a graphical tool to portray income mobility and identify the association between individual movements and initial status which, despite its importance when assessing the social relevance of mobility, is often discounted by aggregate mobility indices. Thu, 01 Mar 2018 10:35:56 GMT http://hdl.handle.net/10993/35082 Title: Kernel-smoothed cumulative distribution function estimation with akdensity <br/> <br/>Author, co-author: Van Kerm, Philippe <br/> <br/>Abstract: In this article, I describe estimation of the kernel-smoothed cumulative distribution function with the user-written package akdensity, with formulas and an example. Thu, 01 Mar 2018 10:30:15 GMT http://hdl.handle.net/10993/35079 Title: Generalized measures of wage differentials <br/> <br/>Author, co-author: van Kerm, Philippe <br/> <br/>Abstract: This paper considers measures of wage differentials not solely determined by mean comparisons but summarizing differences across complete wage distributions. The approach builds on considerations of risk or inequality aversion and on standard expected utility concepts. In an application to the gender pay gap in Luxembourg the disadvantage of women persists according to the proposed measures: lower mean wages for women are not compensated by differences in higher moments of wage distributions (e.g., by less dispersion) at least for realistic assumptions about women preferences toward risk and inequality. The paper also illustrates an original empirical model for wage distributions in the presence of covariates and under endogenous labour market participation. Thu, 01 Mar 2018 10:26:31 GMT http://hdl.handle.net/10993/35074 Title: Space-filling location selection <br/> <br/>Author, co-author: Bia, Michela; Van Kerm, Philippe <br/> <br/>Abstract: This note describes a Stata implementation of a space-filling location selection algorithm. It optimally selects a subset from an array of locations so that the spatial coverage of the array by the selected subset is optimized according to a geometric criterion. Such an algorithm is useful in site selection problems, but also in various non-parametric estimation procedures, e.g. to select (multivariate) knot locations in spline regression analysis. Thu, 01 Mar 2018 10:17:19 GMT http://hdl.handle.net/10993/35073 Title: Decomposing quantile wage gaps: a conditional likelihood approach <br/> <br/>Author, co-author: Van Kerm, Philippe; Choe, Chung; Yu, Seunghee <br/> <br/>Abstract: The paper develops a parametric variant of the Machado–Mata simulation methodology to examine quantile wage differences between groups of workers, with an application to the wage gap between native and foreign workers in Luxembourg. Relying on conditional-likelihood-based ‘parametric quantile regression’ in place of the standard linear quantile regression is parsimonious and cuts computing time drastically with no loss in the accuracy of marginal quantile simulations in our application. We find that the native worker advantage is a concave function of quantile: the advantage is small (possibly negative) for both low and high quantiles, but it is large for the middle half of the quantile range (between the 20th and 70th native wage percentiles). Thu, 01 Mar 2018 10:15:35 GMT http://hdl.handle.net/10993/35070 Title: Accounting for income distribution differences over time and across countries <br/> <br/>Author, co-author: Van Kerm, Philippe Wed, 28 Feb 2018 17:45:40 GMT