Cookie usage policy

The website of the University Carlos III of Madrid use its own cookies and third-party cookies to improve our services by analyzing their browsing habits. By continuing navigation, we understand that it accepts our cookie policy. "Usage rules"

[Close]

Statistics for Data Science

  • Inicio

    Director

    Prof. David Delgado Gómez

    About the Program

    The PhD program in Statistics for Data Science is the natural continuation of the second cycle of graduate studies leading to a master's degree related to Data Science. According to the profile with which the student joins the program, the student may require advanced training courses for the elaboration and presentation of an original research work that will constitute the doctoral thesis.

    Objectives

    The program provides its students with the necessary tools to:        

    • Understand analytical models in Data Science and relate them to their area of knowledge.       
    • Apply advanced knowledge of Statistics for Data Science in the development of methods for the analysis of real problems.
    • Manage at an expert level open programming languages widely used in Data Science, such as R and Python, for the development of statistical analysis.     
    • Develop advanced statistical models for the analysis of real problems involving the prediction of one or several response variables.  
    • Develop statistical models for unsupervised learning.

    Program regulated by RD 99/2011, January 2018

  • ACCESS

    Student profile

    Applicants must have a background in Mathematics, advanced skills in statistical methods and intermediate skills in computer science (programming). Therefore, the preferred profile of access are students with a degree in Statistics, Mathematical Sciences, Physical Sciences, Computer Engineering, Industrial Engineering or Telecommunications Engineering with a master's degree in Data Science. Students with a degree in Data Science complemented with an official master's degree in some area of knowledge are also preferred.

    Admission requirements

    As a general rule, to access the PhD program it is required to hold an official Spanish Bachelor's degree in the field of Statistics, Data Science, Mathematical Sciences, Physical Sciences, Computer Engineering, Industrial Engineering or Telecommunications Engineering and a Master's degree in Data Science for  students with a degree other than Data Science or its equivalents from other EHEA countries.

    Likewise, access is available to candidates in possession of foreign degrees from countries integrated into the European Higher Education Area (EHEA) when the degree can be accredited as level 7 in the European Qualifications Framework (EQF), as long as the aforementioned degree allows access to PhD level studies in the country of expedition; and candidates with a degree which is equivalent to a Spanish Master's degree, obtained in foreign education systems outside the EHEA, as long as the aforementioned degree allows access to PhD level studies in the country of expedition.

    Students with the aforementioned Bachelor's degrees with graduate training in areas other than Data Science can be admitted in the following cases:

    • Students with graduate training in Statistics/Mathematics: The Academic Committee can establish up to a maximum of 6 ECTS of complementary training in advanced subjects related to the line of research the student intends to join.
    • Students with graduate training in Physical Sciences, Computer Engineering, Industrial Engineering or Telecommunications Engineering: The Academic Committee can establish a maximum of 9 ECTS of complementary training in advanced subjects related to the line of research the student intends to join.

    Admission Criteria

    Admission is determined according to the criteria set by the Academic Committee listed below. These are implemented by the Director of the PhD, who reports to the Committee on a regular basis. 

    In the selection and admission process, the Academic Committee will take into account the following aspects:

    • Academic record. Applicant's training (subjects, grades and language skills) in relation to the lines of research of the PhD (60%).
    • Sufficient English skills (minimum B2 or equivalent) (10%).
    • Research experience (publications, conferences, etc.) (15%)
    • Professional experience in relation to Data Science (15%)

    All applications must include a letter of motivation from the candidate expressing their research interests in relation to the lines of research of the PhD.

    Applicants must submit letter(s) of recommendation from academic/scientific experts from other institutions.

    In the comparative evaluation of the applications for admission, priority will be given to applications fitting the preferred profile. The rest of profiles will be subsequently considered if there are places available.

    The Academic Committee may request a personal interview for clarification of the details provided in the application. This interview is not an evaluation factor with a specific weight but an additional instrument to clarify the suitability of the profile and the student's motivation.

    Seats available for the academic year: 10

  • FACULTY
  • TRAINING

    Concurrently with the doctoral research work, PhD candidates are required to follow a training program to improve their research skills and ensure the scientific quality of their research. This program is structured on the following elements.

    Specific training

    • Specific Research Seminars

    Mandatory. 20 hours throughout the PhD studies

    Seminars organized by the Department of Statistics and related to the lines of research of the PhD, taught by expert speakers from international, mostly from outside Spain.

    For evaluation, the PhD candidate will provide proof of attendance and a report on the seminar content (critical analysis and aspects of research with direct impact on the student's research). The student will receive a certificate of completion that must be registered on their document of activities. 

    • Workshop on a current topic in Data Science

    Mandatory. 20-30 hours throughout the PhD studies (attendance to at least two workshops during the PhD studies).

    Annual workshop of 10-15 hours of duration, lectured during one week. Possible topics: probabilistic graphical models, network analysis, Bayesian variational methods or Data Science in health.

    For evaluation, the PhD candidate will provide proof of attendance and a report in which the seminar contents are applied (critical analysis and aspects of research with direct impact on the student's research). The student will receive a certificate of completion that must be registered on their document of activities. 

    • Presentation of two working papers

    Mandatory. 200 hours throughout the PhD.

    Two working papers that will be published in open access with the collaboration of the Library Service.

    Each working paper will be evaluated by a committee of experts from the Department. This evaluation will not be carried out if the paper is accepted by a peer-reviewed journal.

    • Mid-doctorate and pre-thesis defense presentations

    Mandatory. 40 hours.

    The PhD candidate will give two presentations about their research.

    The first presentation will be carried out mid-doctorate to ensure the correct development of the doctoral studies.

    The second presentation will be carried out prior to the thesis defense to ensure that the thesis has enough quality to initiate the process.

    For the mid-doctorate presentation, a committee of the three researchers most closely related to the research topic will be constituted. In case of unforeseen circumstances, this committee will establish the guidelines and the maximum time to correct deficiencies.

    For the pre-thesis defense presentation. a committee of the three researchers most closely related to the research topic will also be constituted. In case of minor deficiencies, the PhD candidate must correct them before the official deposit of the thesis. If there are greater deficiencies, the process will be suspended until those are corrected.

    • Research Stay

    Optional. 45-90 days.

    Research stay, preferably international, in a prestigious research center in the field of the thesis. The aim of the stay is to complete the training in an international expert group and promoting the research.    

    If the stay is longer than 3 months and the work program has been completed successfully, this may be considered as a merit to obtain the International PhD distinction.

    Students will provide the proof of completion granted by the host center once the stay is over.

    • Lectures by the PhD candidate

    Optional. 10 hours per academic year.

    Presentation of the PhD candidate's work in international conferences. Students are expected to present their research in at least two international conferences throughout the PhD. 

    Additionally, the Spanish Young Statisticians and Operational Researchers Meeting (SYSORM), where students can share knowledge, experiences and concerns, is being held in recent years.

    Students must provide the accreditation from the conference organizing committee.

    Research skills training

    Optional. Up to 2 credits throughout the PhD. 

    Research skills training is focused on abilities common to all disciplines for the development of scientific and educational skills and the improvement of the professional career. This training consists of different activities (short courses, seminars, etc.), which can be recommended by the Academic Committee of the program.

    Further information:

  • RESEARCH
    • Lines of research
      • Statistical and quantitative methods for Data Science
      • Operation research for Data Science
    • Scientific results

      A sample of relevant publications from doctoral theses of the Program will be listed in this section.

    • Scientific publications

      A sample of relevant faculty publications are listed below:

      • Perez-Santalla, R.; Carrión, M.; Ruiz, C. "Optimal pricing for electricity retailers based on data-driven consumers’ price-response". TOP. 2022. ISNN: 1134-5764
        DOI: https://doi.org/10.1007/s11750-022-00622-8
      • Lee D.; Durbán, M.; Ayma, D.; Van de Kassteele, J. "Modeling latent spatio-temporal disease incidence using penalized composite link models". Plos One. 2022. ISSN: 1932-6203
        DOI: https://doi.org/10.1371/journal.pone.0263711
      • Mendez-Cevieta, A.; Aguilera-Morillo, C.; Lillo, R. E. "Fast partial quantile regression. Chemometrics and Intelligent Laboratory Systems". 2022. ISSN: 0169-7439
        DOI: https://doi.org/10.1016/j.chemolab.2022.104533
      • Abate, A. G.; Riccardi, R.; Ruiz, C. "Contract design in electricity markets with high penetration of renewables: A two-stage approach". Omega. 2022. ISNN: 0305-0483
        DOI: https://doi.org/10.1016/j.omega.2022.102666
      • Llorente, F.; Martino, L.; Read, J.; Delgado-Gómez, D. "Optimality in noisy importance sampling. Signal Processing". 2022. ISSN: 0165-1684
        DOI: https://doi.org/10.1016/j.sigpro.2022.108455
      • Alemán-Gómez, Y.; Arribas-Gil, A.; Desco, M.; Elias, A.; Romo, J. Depthgram. "Visualizing outliers in high-dimensional functional data with application to fMRI data exploration". Statistics in Medicine. 2022. ISSN: 0277-6715
        DOI: https://doi.org/10.1002/sim.9342
      • Cascos, I., López-Díaz, C. & López-Díaz, M. "A stochastic order for interval valued random mappings and applications". Applied Mathematical Modelling 107, 429–440 (2022)
        DOI: https://doi.org/10.1016/j.apm.2022.02.039
      • Carrizosa, E.; Guerrero, V.; Morales, D.R. "On mathematical optimization for clustering categories in contingency tables". Advances in Data Analysis and Classification. 2022. ISSN: 1862-5347
        DOI: https://doi.org/10.1007/s11634-022-00508-4
      • Grané, A; Albarrán, I.; Guo, Q. "Visualizing Health and Well-Being Inequalities Among Older Europeans". Social Indicators Research. 2021. ISSN: 0303-8300
        DOI: https://doi.org/10.1007/s11205-021-02621-x
      • Cascos, I. "Simultaneous monitoring of origin and scale in left-bounded processes via depth". AStA Advances in Statistical Analysis. 2021. ISSN: 1863-818X
        DOI: https://doi.org/10.1007/s10182-021-00401-z
      • Cabras, S. "A Bayesian-Deep Learning Model for Estimating COVID-19 Evolution in Spain". Mathematics. 2021. ISNN: 2227-7390
        DOI: https://doi.org/10.3390/math9222921
      • Elías, A.; Jiménez, R.; Shang, H. L. "On projection methods for functional time series forecasting". Journal of Multivariate Analysis. 2021. ISSN: 0047-259X
        DOI: https://doi.org/10.1016/j.jmva.2021.104890
      • Alonso, A. M.; D’Urso, P.; Gamboa, C.; Guerrero, V. "Cophenetic-based fuzzy clustering of time series by linear dependency". International Journal of Approximate Reasoning. 2021. ISSN: 0888-613X
        DOI: https://doi.org/10.1016/j.ijar.2021.07.006
      • Gómez de Mariscal, E.; Guerrero, V.; Sneider, A.; Jayatilaka, H.; Wirth P. D.; Muñoz-Barrutia, A. "Use of the p-values as a size-dependent function to address practical differences when analyzing large datasets". Scientific Reports. 2021. ISSN: 2045-2322
        DOI: https://doi.org/10.1038/s41598-021-00199-5
      • Quijano, J.; Liberatore, F.; Rodriguez-Lorenzo, G.;Lillo, R. E.; González-Alvarez, J. "A twist in Intimate Partner Violence Risk Assessment Tools: Gauging the contribution of exogenous and historical variables". Knowledge-Based Systems. 2021. ISSN: 0950-7051
        DOI: https://doi.org/10.1016/j.knosys.2021.107586
      • Morala, P.; Cifuentes, J.;Lillo, R. E.; Ucar, I. "Towards a mathematical framework to inform neural network modelling via polynomial regression". Neural Networks. 2021. ISNN: 0893-6080
        DOI: https://doi.org/10.1016/j.neunet.2021.04.036
      • Poncela, P.; Ruiz, E.; Miranda, K. "Factor extraction using Kalman filter and smoothing: This is not just another survey". International Journal of Forecasting. 2021. ISNN: 0169-2070
        DOI: https://doi.org/10.1016/j.ijforecast.2021.01.027
      • Cabras, S.; Castellanos, M. E.; Ratmann, O. "Goodness of fit for models with intractable likelihood". Test. 2021. ISNN: 1133-0686
        DOI: https://doi.org/10.1007/s11749-020-00747-7
      • Liu, L.; Martín-Barragan,  B.; Prieto, J. "A projection multi-objective SVM method for multi-class classification". Computers & Industrial Engineering. 2021. ISSN: 0360-8352
        DOI: https://doi.org/10.1016/j.cie.2021.107425
      • Cascos, I.; Ochoa, M. "Expectile depth: Theory and computation for bivariate datasets". Revista: Journal of Multivariate Analysis. 2021. ISSN: 0047-259X
        DOI: https://doi.org/10.1016/j.jmva.2021.104757
      • Laria, J. C.; Delgado-Gómez, D.; Peñuelas-Calvo, I.; Baca-García, E.; Lillo, R. "Accurate Prediction of Children's ADHD Severity Using Family Burden Information: A Neural Lasso Approach". Frontiers in Computational Neuroscience. 2021. ISSN: 1662-5188
        DOI: https://doi.org/10.3389/fncom.2021.674028
      • Strzalkowska-Kominiak, E.; Romo, J. "Censored functional data for incomplete follow-up studies". Statistics in Medicine. 2021. ISSN: 0277-6715
        DOI: https://doi.org/10.1002/sim.8930
      • Meilán-Vila, A; Fernández-Casas,R.; Crujeiras, R.M; Francisco-Fernández, M. "A computational validation for nonparametric assessment of spatial trends". Computational Statistics. 2021. ISSN: 0943-4062
        DOI: https://doi.org/10.1007/s00180-021-01108-0
      • Oliveira F. S.; Ruiz, C. "Analysis of futures and spot electricity markets under risk aversion". European Journal of Operational Research. 2020. ISSN: 0377-2217
        DOI: https://doi.org/10.1016/j.ejor.2020.10.005
      • García-Portugués, E.; Álvarez Liebana, J., Álvarez-Pérez, G.; González-Manteiga, W. "A goodness-of-fit test for the functional linear model with functional response". Scandinavian Journal of Statistics. 2020. ISSN: 0303-6898
        DOI: https://doi.org/10.1111/sjos.12486
  • THESIS

    Preparation of the thesis

    Please check the Guide with recommendations that the University Library has accessible on its website.

    You must always follow the guidelines of your thesis advisor, the guidelines of your doctoral program, and the regulations of the Doctoral School.

    Thesis defense

    The doctoral thesis consists of an original research work developed by the Ph.D. candidate student in the field of knowledge of the program that enables the student for autonomous work in the field of R+D+ i.

    Universidad Carlos III de Madrid and the Doctoral School establishes the follow-up procedures to guarantee the quality of the doctorate's training and supervision. It also facilitates the procedures for the proper evaluation and defense of the doctoral thesis.

    Further information:

    Requirements for thesis defense

    Give a pre-defense seminar before a panel of three department researchers related to the subject of the thesis. Students will present the most relevant outcomes of their research, which must have resulted in at least two working papers previously evaluated and approved by a committee of experts.

    The student will provide the members of the panel with a copy of the thesis two weeks prior to the seminar. After the presentation and questions, the panel will give their assessment:

    • Approved. The panel considers that the thesis meets the quality requirements and authorizes the initiation of the defense process.
    • Approved with minor modifications. The panel considers that the thesis meets the quality requirements but minor structural, typographical, or clarifying corrections are necessary. The defense is authorized once these recommendations have been included in the thesis.
    • Requirement of major modifications. The panel considers that the thesis does not meet the quality requirements and that it should not be deposited (e.g., inconsistencies or lack of relevant experiments). The student must correct these weaknesses and undergo a new pre-defense process no sooner than six months.
  • QUALITY

    GENERAL INFORMATION ABOUT THE PH.D.

    Implementation Year: 2023-2024

    QUALITY ASSURANCE

    The Academic Committee of the Ph.D. complies with the Internal Quality Assurance System  SGIC-UC3M quality of the Ph.D. Program and produces the Degree Reports ("Memoria Académica de Titulación").

    • Academic Committee
    • Reports from the Quality Assurance Committee (Restricted access) (Available soon)

    QUALITY INDICATORS 

    COMPLAINTS AND SUGGESTIONS

  • CONTACT
    Bienvenida Universidad Carlos III de Madrid

    Doctoral School Office | Leganés Campus

    Rey Pastor Building, Office 3.0.B.08
    Avenida de la Universidad, 30
    28911 Leganés (Madrid)

    Contact