Ana Pérez-González

Nonparametric Mean Estimation with Missing Data

Abstract Missing data occur in most applied statistical analysis. The need to estimate the conditional or unconditional mean of a variable when some of its observations are missing is very frequent. In this article we study the effect of missing observations on the response variable in the estimation of a multivariate regression function. This effect is also considered in the estimation of the marginal mean. Following the research of Chu and Cheng [Chu, C. K., Cheng, P. E. (1995). Nonparametric regression estimation with missing data. J. Statist. Planning Inference 48:85–99] for the univariate case, we propose three non-parametric estimators of the regression function based on the Multivariate Local Linear Smoother [see Ruppert, D., Wand, M. P. (1994). Multivariate locally weighted least squares regression. Ann. Statist. 22(3):1346–1370]. The first consists of using only complete observations; the other two use simple or multiple imputation techniques respectively to complete the sample. The behavior in function of the Asymptotic Mean Squared Error (AMSE) is studied for the estimators. A method for obtaining optimal estimated bandwidth matrices based on the Bootstrap resampling mechanism is proposed.

Estimation of the marginal location under a partially linear model with missing responses

In this paper, we consider a semiparametric partially linear regression model where there are missing data in the response. We propose robust Fisher-consistent estimators for the regression parameter, for the regression function and for the marginal location parameter of the response variable. A robust cross-validation method is briefly discussed, although, from our numerical results, the marginal estimators seem not to be sensitive to the bandwidth parameter. Finally, a Monte Carlo study is carried out to compare the performance of the robust proposed estimators among themselves and also with the classical ones, for normal and contaminated samples, under different missing data models. An example based on a real data set is also discussed.

Model checks for nonparametric regression with missing data: a comparative study

ABSTRACT This paper analyses the behaviour of the goodness-of-fit tests for regression models. To this end, it uses statistics based on an estimation of the integrated regression function with missing observations either in the response variable or in some of the covariates. It proposes several versions of one empirical process, constructed from a previous estimation, that uses only the complete observations or replaces the missing observations with imputed values. In the case of missing covariates, a link model is used to fill the missing observations with other complete covariates. In all the situations, Bootstrap methodology is used to calibrate the distribution of the test statistics. A broad simulation study compares the different procedures based on empirical regression methodology, with smoothed tests previously studied in the literature. The comparison reflects the effect of the correlation between the covariates in the tests based on the imputed sample for missing covariates. In addition, the paper proposes a computational binning strategy to evaluate the tests based on an empirical process for large data sets. Finally, two applications to real data illustrate the performance of the tests.

Measuring the efficiency of the Spanish Olympic Sports Federations

ABSTRACT Research question: The most common result in the analysis of efficiency in multisport competitions like the Olympic Games using the data envelopment analysis (DEA) is a ranking of the participating countries. However, this approach does not consider the role played by sports federations, who administer sports at the national level, nor the intermediate stages needed to achieve sporting success. The objective of the present study is to analyze the relative efficiency of the Spanish Olympic Sports Federations, taking into account what occurs within the black box of these organizations. Research methods: The relative efficiency of Spanish Olympic Sports Federations during the last three years (2010, 2011 and 2012) of the last Olympic cycle is measured by employing a relational network DEA model. This type of model can simultaneously calculate the efficiency of the system and its different stages to represent the eventual transformation of public and private financing into sport results. Results and findings: The findings reveal that during the last Olympic cycle, the efficiency of the Federations in supporting the development of high-level athletes was greater than the effectiveness of those individuals when participated in the main international competitions. Moreover, the differences in efficiency distributions tend to disappear in the last year of the Olympic cycle. Implications: These empirical findings provide the managers with useful information regarding the origins of system inefficiencies and their distribution throughout the Olympic cycle. This information can help them improve national sport programs.

Nonparametric variance function estimation with missing data

In this paper, a fixed design regression model where the errors follow a strictly stationary process is considered. In this model the conditional mean function and the conditional variance function are unknown curves. Correlated errors when observations are missing in the response variable are assumed. Four nonparametric estimators of the conditional variance function based on local polynomial fitting are proposed. Expressions of the asymptotic bias and variance of these estimators are obtained. A simulation study illustrates the behavior of the proposed estimators.

Cultural differences, language attitudes and tourist satisfaction: a study in the Barcelona hotel sector

ABSTRACT In most service activities, customer satisfaction depends largely on the direct interaction with service providers. In the case of tourism, this interaction often occurs between people from different countries and whose mother tongues are different. In this context, concepts such as cultural proximity, linguistic accommodation, expectations and language attitudes enrich the analysis of tourist satisfaction. This study uses an interdisciplinary approach that integrates cultural and linguistic elements in the analysis of tourist evaluations in the hotel sector. In particular, the technique of Content Analysis is applied to comments made by hotel guests on Booking.com to determine the extent to which cultural differences, understood in terms of country of origin, influence two relevant aspects of the tourist destination experience: their language attitudes and level of satisfaction with the service received. The findings confirm that the country of origin influences guests’ evaluations and the choice of the language in which the experience is assessed, and reveal that comments concerning language reflect this influence. In particular, linguistic experiences seem to be more important for tourists who emphasise the role of their mother tongue (Italians and French) than for tourists more open to the use of other languages (German and Portuguese).