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


Mariano Matilla-García has a total of 65 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Specification Testing in Cross-Sectional Regression with Mixed Regressors: A Tuning-Free Nonparametric Approach

    by Mariano Matilla-Garcia, Manuel Ruiz

    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.

  • Ordinal patterns based testing of spatial independence in irregular spatial structures

    by Giorgio Micali, David Garnés-Galindo, Mariano Matilla-García, Manuel Ruiz-Marín

    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

    Specification Testing in Cross-Sectional Regression with Mixed Regressors: A Tuning-Free Nonparametric Approach

    by Mariano Matilla-Garcia, Manuel Ruiz

    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

    Improving Home Insurance Ratemaking with Geographically Weighted Poisson Regression (GWPR) Model: Assessing Water Damage Risk

    by Maria-Victoria Rivas-Lopez, Mariano Matilla-García, Roman Minguez-Salido, Miguel Angel Bravo-Ovalle

    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

    Multibenchmark reality checks

    by Ignacio Arbués, Mariano Matilla-García

  • Open Access

    Economic burden of Cardiac Arrest in Spain: analyzing healthcare costs drivers and treatment strategies cost-effectiveness

    by Mariano Matilla-García, Paloma Ubeda Molla, Fernando Sánchez Martínez, Albert Ariza-Solé, Rocío Gómez-López, Esteban López de Sá, Ricard Ferrer

    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.


  • Non-parametric analysis of serial dependence in time series using ordinal patterns

    by Christian H. Weiß, Manuel Ruiz Marín, Karsten Keller, Mariano Matilla-García


  • Spatial partial causality

    by Marcos Herrera-Gomez, Mariano Matilla-Garcia, Manuel Ruiz-Marin

  • Open Access

    Information Theory and Symbolic Analysis: Theory and Applications

    by Mariano Matilla-García, Manuel Ruiz Marín

    Symbolic analysis has been developed and used successfully in very diverse fields