Mariano Matilla-García
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Mariano Matilla-García has a total of 65 co-authors. Here are the top co-authors by number of shared publications:
- Manuel Ruiz Marín
Departamento de Métodos Cuantitativos, Universidad Politécnica de Cartagena 2 , 30201 Cartagena
28 shared publications - Fernando A. López
Universidad Politécnica de Cartagena
4 shared publications - Francisco Jose Vazquez
Departamento de Análisis Económico: Economía Cuantitativa, Universidad Autónoma de Madrid, Madrid, Spain
3 shared publications - Jesús Mur
University of Zaragoza
3 shared publications - José Antonio García Cordoba
Departamento de Métodos Cuantitativos, Ciencias Jurídicas y Lenguas Modernas, Universidad Politécnica de Cartagena, Calle Real 3, Edif. CIM, 30202 Cartagena, Spain
3 shared publications Jesús Mur
Campus Rio Ebro University of Zaragoza
2 shared publications
Latest Publications
- Open Access
- 2 September 2026
- Published by Elsevier in SSRN Electronic Journal
Specification Testing in Cross-Sectional Regression with Mixed Regressors: A Tuning-Free Nonparametric Approach
We propose a class of nonparametric specification tests for cross-sectional regression models based on ordinal-pattern representations of regressors and errors. The method represents the joint distribution of regressors and errors through a finite set of ordinal patterns, transforming specification testing into a problem defined on a fixed probability simplex. The resulting procedures are fully automatic: beyond specifying the regression model, the user need not make any methodological choice. The finite-alphabet structure yields a transparent large-sample theory whose limit distribution follows from classical central limit arguments on a fixed-dimensional probability simplex, and supports resampling-based inference adapted to both independence and mean exogeneity. The method applies naturally to continuous, discrete, or mixed regressors without modification. Monte Carlo evidence shows reliable size control and strong power against nonlinear and interaction-based alternatives, including designs with discrete covariates where conventional nonparametric tests are difficult to implement.
- 2 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSpecification Testing in Cross-Sectional Regression with Mixed Regressors: A Tuning-Free Nonparametric Approach
We propose a class of nonparametric specification tests for cross-sectional regression models based on ordinal-pattern representations of regressors and errors. The method represents the joint distribution of regressors and errors through a finite set of ordinal patterns, transforming specification testing into a problem defined on a fixed probability simplex. The resulting procedures are fully automatic: beyond specifying the regression model, the user need not make any methodological choice. The finite-alphabet structure yields a transparent large-sample theory whose limit distribution follows from classical central limit arguments on a fixed-dimensional probability simplex, and supports resampling-based inference adapted to both independence and mean exogeneity. The method applies naturally to continuous, discrete, or mixed regressors without modification. Monte Carlo evidence shows reliable size control and strong power against nonlinear and interaction-based alternatives, including designs with discrete covariates where conventional nonparametric tests are difficult to implement.
- 1 June 2026
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 36 (6)
Ordinal patterns based testing of spatial independence in irregular spatial structures
We propose a nonparametric test of spatial independence for data observed on irregular, non-lattice point clouds Vn⊂R2. For each location v∈Vn, we encode the local spatial configuration through the ordinal pattern of the m nearest-neighbor observations, obtaining a symbolic representation that is invariant under strictly monotone transformations and robust to outliers. Under the null hypothesis of spatial independence, the local ordinal patterns are i.i.d. and uniformly distributed over the symmetric group Sm, regardless of the unknown marginal distribution F. We exploit this characterization to construct a test statistic Ln based on the additive log-ratio transformation of the empirical ordinal-pattern frequencies. Invoking a central limit theorem for graph-dependent processes under a graph-based α-mixing condition, we establish that Ln converges in distribution to a χm!−12 random variable, yielding an asymptotically pivotal procedure with no nuisance parameters. An extensive Monte Carlo study confirms that the χm!−12 approximation is accurate already at moderate sample sizes, that the test controls size at the nominal level, and that power increases monotonically with the strength of spatial dependence. Notably, the test detects dependence in both linear and nonlinearly transformed spatial autoregressive models, illustrating the robustness that is characteristic of ordinal-pattern methods. Our framework extends the spatial ordinal-pattern testing paradigm from regular lattices to general spatial supports, opening the door to ordinal-pattern inference in the many applied settings where observations are irregularly located.
- 1 June 2026
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 36 (6)
Ordinal patterns based testing of spatial independence in irregular spatial structures
We propose a nonparametric test of spatial independence for data observed on irregular, non-lattice point clouds Vn⊂R2. For each location v∈Vn, we encode the local spatial configuration through the ordinal pattern of the m nearest-neighbor observations, obtaining a symbolic representation that is invariant under strictly monotone transformations and robust to outliers. Under the null hypothesis of spatial independence, the local ordinal patterns are i.i.d. and uniformly distributed over the symmetric group Sm, regardless of the unknown marginal distribution F. We exploit this characterization to construct a test statistic Ln based on the additive log-ratio transformation of the empirical ordinal-pattern frequencies. Invoking a central limit theorem for graph-dependent processes under a graph-based α-mixing condition, we establish that Ln converges in distribution to a χm!−12 random variable, yielding an asymptotically pivotal procedure with no nuisance parameters. An extensive Monte Carlo study confirms that the χm!−12 approximation is accurate already at moderate sample sizes, that the test controls size at the nominal level, and that power increases monotonically with the strength of spatial dependence. Notably, the test detects dependence in both linear and nonlinearly transformed spatial autoregressive models, illustrating the robustness that is characteristic of ordinal-pattern methods. Our framework extends the spatial ordinal-pattern testing paradigm from regular lattices to general spatial supports, opening the door to ordinal-pattern inference in the many applied settings where observations are irregularly located.
- Open Access
- 17 March 2026
- Published by Elsevier in SSRN Electronic Journal
Specification Testing in Cross-Sectional Regression with Mixed Regressors: A Tuning-Free Nonparametric Approach
We propose a class of nonparametric specification tests for cross-sectional regression models based on ordinal-pattern representations of regressors and errors. The method represents the joint distribution of regressors and errors through a finite set of ordinal patterns, transforming specification testing into a problem defined on a fixed probability simplex. The resulting procedures are fully automatic: beyond specifying the regression model, the user need not make any methodological choice. The finite-alphabet structure yields a transparent large-sample theory whose limit distribution follows from classical central limit arguments on a fixed-dimensional probability simplex, and supports resampling-based inference adapted to both independence and mean exogeneity. The method applies naturally to continuous, discrete, or mixed regressors without modification. Monte Carlo evidence shows reliable size control and strong power against nonlinear and interaction-based alternatives, including designs with discrete covariates where conventional nonparametric tests are difficult to implement.
- 17 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSpecification Testing in Cross-Sectional Regression with Mixed Regressors: A Tuning-Free Nonparametric Approach
We propose a class of nonparametric specification tests for cross-sectional regression models based on ordinal-pattern representations of regressors and errors. The method represents the joint distribution of regressors and errors through a finite set of ordinal patterns, transforming specification testing into a problem defined on a fixed probability simplex. The resulting procedures are fully automatic: beyond specifying the regression model, the user need not make any methodological choice. The finite-alphabet structure yields a transparent large-sample theory whose limit distribution follows from classical central limit arguments on a fixed-dimensional probability simplex, and supports resampling-based inference adapted to both independence and mean exogeneity. The method applies naturally to continuous, discrete, or mixed regressors without modification. Monte Carlo evidence shows reliable size control and strong power against nonlinear and interaction-based alternatives, including designs with discrete covariates where conventional nonparametric tests are difficult to implement.
- Open Access
- 11 February 2025
- Published by Springer Nature in Applied Spatial Analysis and Policy
- Vol. 18 (1) , 1-32
Improving Home Insurance Ratemaking with Geographically Weighted Poisson Regression (GWPR) Model: Assessing Water Damage Risk
This study examines the consideration of spatial heterogeneity in the development of Home Insurance rates, specifically focusing on water damage throughout Spain. This focus arises from the need to establish a methodology that not only improves ratemaking procedures for water damage but also acknowledges the potential impacts of climate change, allowing differentiation in the effect of variables such as rainfall depending on the location and frequency of water claims. By using the GWPR model, spatial heterogeneity is taken into account and the ratemaking process is enhanced by identifying spatial clusters related to the frequency of water damage claims. Moreover, an empirical development has been carried out employing a database of home insurance data for water coverage in the Spanish territory. The variables selected in this process are not only associated with weather, but also with characteristics of the policies, housing, and socio-economic conditions of the policyholders.
- 11 February 2025
- Published by Springer Nature in Applied Spatial Analysis and Policy
- Vol. 18 (1) , 1-32
Open AccessImproving Home Insurance Ratemaking with Geographically Weighted Poisson Regression (GWPR) Model: Assessing Water Damage Risk
This study examines the consideration of spatial heterogeneity in the development of Home Insurance rates, specifically focusing on water damage throughout Spain. This focus arises from the need to establish a methodology that not only improves ratemaking procedures for water damage but also acknowledges the potential impacts of climate change, allowing differentiation in the effect of variables such as rainfall depending on the location and frequency of water claims. By using the GWPR model, spatial heterogeneity is taken into account and the ratemaking process is enhanced by identifying spatial clusters related to the frequency of water damage claims. Moreover, an empirical development has been carried out employing a database of home insurance data for water coverage in the Spanish territory. The variables selected in this process are not only associated with weather, but also with characteristics of the policies, housing, and socio-economic conditions of the policyholders.
- Open Access
- 14 August 2024
- Published by Elsevier in Economic Modelling
- Vol. 140
Multibenchmark reality checks
- 14 August 2024
- Published by Elsevier in Economic Modelling
- Vol. 140
Open AccessMultibenchmark reality checks
- 5 August 2024
- Published by Taylor & Francis in Applied Economics
- Vol. 57 (40) , 6258-6273
An empirical analysis of the impacts of competition on procurement costs
- 5 August 2024
- Published by Taylor & Francis in Applied Economics
- Vol. 57 (40) , 6258-6273
An empirical analysis of the impacts of competition on procurement costs
- Open Access
- 7 November 2023
- Published by Springer Nature in BMC Health Services Research
- Vol. 23 (1) , 1-9
Economic burden of Cardiac Arrest in Spain: analyzing healthcare costs drivers and treatment strategies cost-effectiveness
Cardiac arrest is a major public health issue in Europe. Cardiac arrest seems to be associated with a large socioeconomic burden in terms of resource utilization and health care costs. The aim of this study is the analysis of the economic burden of cardiac arrest in Spain and a cost-effectiveness analysis of the key intervention identified, especially in relation to neurological outcome at discharge. The data comes from the information provided by 115 intensive care and cardiology units from Spain, including information on the care of patients with out-of-hospital cardiac arrest who had a return of spontaneous circulation. The information reported by theses 115 units was collected by a nationwide survey conducted between March and September 2020. Along with number of patients (2631), we also collect information about the structure of the units, temperature management, and prognostication assessments. In this study we analyze the potential association of several factors with neurological outcome at discharge, and the cost associated with the different factors. The cost-effectiveness of using servo-control for temperature management is analyzed by means of a decision model, based on the results of the survey and data collected in the literature, for a one-year and a lifetime time horizon. A total of 109 cardiology units provided results on neurological outcome at discharge as evaluated with the cerebral performance category (CPC). The most relevant factor associated with neurological outcome at discharge was ‘servo-control use’, showing a 12.8% decrease in patients with unfavorable neurological outcomes (i.e., CPC3-4 vs. CPC1-2). The total cost per patient (2020 Euros) was €73,502. Only “servo-control use” was associated with an increased mean total cost per hospital. Patients treated with servo-control for temperature management gained in the short term (1 year) an average of 0.039 QALYs over those who were treated with other methods at an increased cost of €70.8, leading to an incremental cost-effectiveness ratio of 1,808 euros. For a lifetime time horizon, the use of servo-control is both more effective and less costly than the alternative. Our results suggest the implementation of servo-control techniques in all the units that are involved in managing the cardiac arrest patient from admission until discharge from hospital to minimize the neurological damage to patients and to reduce costs to the health and social security system.
- 7 November 2023
- Published by Springer Nature in BMC Health Services Research
- Vol. 23 (1) , 1-9
Open AccessEconomic burden of Cardiac Arrest in Spain: analyzing healthcare costs drivers and treatment strategies cost-effectiveness
Cardiac arrest is a major public health issue in Europe. Cardiac arrest seems to be associated with a large socioeconomic burden in terms of resource utilization and health care costs. The aim of this study is the analysis of the economic burden of cardiac arrest in Spain and a cost-effectiveness analysis of the key intervention identified, especially in relation to neurological outcome at discharge. The data comes from the information provided by 115 intensive care and cardiology units from Spain, including information on the care of patients with out-of-hospital cardiac arrest who had a return of spontaneous circulation. The information reported by theses 115 units was collected by a nationwide survey conducted between March and September 2020. Along with number of patients (2631), we also collect information about the structure of the units, temperature management, and prognostication assessments. In this study we analyze the potential association of several factors with neurological outcome at discharge, and the cost associated with the different factors. The cost-effectiveness of using servo-control for temperature management is analyzed by means of a decision model, based on the results of the survey and data collected in the literature, for a one-year and a lifetime time horizon. A total of 109 cardiology units provided results on neurological outcome at discharge as evaluated with the cerebral performance category (CPC). The most relevant factor associated with neurological outcome at discharge was ‘servo-control use’, showing a 12.8% decrease in patients with unfavorable neurological outcomes (i.e., CPC3-4 vs. CPC1-2). The total cost per patient (2020 Euros) was €73,502. Only “servo-control use” was associated with an increased mean total cost per hospital. Patients treated with servo-control for temperature management gained in the short term (1 year) an average of 0.039 QALYs over those who were treated with other methods at an increased cost of €70.8, leading to an incremental cost-effectiveness ratio of 1,808 euros. For a lifetime time horizon, the use of servo-control is both more effective and less costly than the alternative. Our results suggest the implementation of servo-control techniques in all the units that are involved in managing the cardiac arrest patient from admission until discharge from hospital to minimize the neurological damage to patients and to reduce costs to the health and social security system.
- 1 April 2022
- Published by Elsevier in Computational Statistics & Data Analysis
- Vol. 168
Non-parametric analysis of serial dependence in time series using ordinal patterns
- 1 April 2022
- Published by Elsevier in Computational Statistics & Data Analysis
- Vol. 168
Non-parametric analysis of serial dependence in time series using ordinal patterns
- 20 December 2021
- Published by Springer Nature in The European Physical Journal Special Topics
- Vol. 231 (9) , 1735-1739
Spatial partial causality
- 20 December 2021
- Published by Springer Nature in The European Physical Journal Special Topics
- Vol. 231 (9) , 1735-1739
Spatial partial causality
- Open Access
Information Theory and Symbolic Analysis: Theory and Applications
Symbolic analysis has been developed and used successfully in very diverse fields
Open AccessInformation Theory and Symbolic Analysis: Theory and Applications
Symbolic analysis has been developed and used successfully in very diverse fields