| Code | SD-Inf-G2 |
|---|---|
| Organizational unit | Doctoral School of Exact Sciences |
| Area/discipline | Computer and information sciences |
| Form of studies | Full-time |
| Level of education | Third cycle |
| Language(s) of instruction | English, Polish |
| Admission limit | 1 |
| Duration | 8 semesters |
| Recruitment committee address | Szkoła Doktorska Nauk Ścisłych, ul. Uniwersytetu Poznańskiego 8, Poznań mgr Agnieszka Kurzajak |
| Office opening hours | 9.00-14.00 |
| WWW address | https://amu.edu.pl/doktoranci/szkola-doktorska/rekrutacja/rekrutacja-20252026 |
| Required document | |
| Ask a question | |
Additional recruitment to the Doctoral School of Exact Sciences beyond the limit of places
Project description:
Functional data analysis (FDA) has become a significant tool in statistics, especially effective when data are densely sampled. It analyzes data that consist of or can be represented by functions, curves, or surfaces. The development of highly accurate measurement instruments allows us to observe such data in many fields of science, such as biology, chemometrics, economics, medicine, and meteorology. For example, when patients wear a device that automatically measures blood pressure and heart rate at regular intervals over a 24-hour period, the output is trajectories of these variables, which can be modeled as functional data. When the frequency of repeated observations over time increases, functional data analysis has significant advantages over traditional statistical data analysis methods because it does not assume specific dependency structures or require evenly spaced time points at which features are measured. This flexibility can lead to better conclusions and allows additional information to be extracted from the data. Despite methodological advances, there is still a need to construct efficient methods for the analysis of functional data to cope with challenges such as missing data, non-normality of the distribution, and heteroscedasticity of the covariance function. The project aims to develop new statistical methods for both univariate and multivariate functional analysis of variance or covariance with repeated measures that are not limited by the assumptions of traditional models. These methods will be constructed to provide reliable results even in difficult cases such as small sample sizes and partially or completely missing observations. We will also consider the presence of covariance factors in different factorial designs and test complex statistical hypotheses involving main and interaction effects in both crossed and hierarchical models, without assuming a specific probability distribution. The construction of statistical tests in these models will rely on both traditional and novel estimators, such as the sample mean and covariance functions, as well as wavelet-based estimators. We will pay special attention to imputation and statistical methods that enable effective inference in the presence of missing data. Such missing data can have different characteristics and causes. The first example is the absence of the entire trajectory, which can occur when the patient had their blood pressure and heart rate measured for the whole of the day before the start of treatment but did not measure it again after its completion. The second example results in so-called partially observed functional data, when the patient turns off the measuring device for several hours, which causes a lack of measurement for a significant part of the day. The project provides for a comprehensive study of the theoretical properties of the constructed statistical procedures and their effectiveness in practice. We plan to use intensive simulation studies to assess their finite-sample properties. The results will be integrated with practical statistical tools in the form of efficient software that will find application in real research problems, especially in medicine. Example applications include the analysis of biomedical and biostatistical data, where missing data are common and accounting for covariates and the functional structure of repeated measures is crucial for correct inference. The project collaboration aims not only to develop methods for the analysis of functional data, focusing on statistical tests and confidence regions, but also to prepare guidelines for applying these methods to biostatistical problems. This comprehensive approach aims to bridge the gap between advanced statistical theory and real-world applications, making it a significant step forward in the field of functional data analysis.
Principal investigator: prof. UAM dr hab. Łukasz Smaga - NCN 2025/07/Y/ST6/00001
Recruitment
Selection Committee
1. prof. UAM dr hab, Łukasz Smaga -chairman
2. prof. UAM dr hab. Joanna Berlińska
3. prof. UAM dr hab. Tomasz Górecki
4. prof. UAM dr hab. Michał Hanćkowiak
5. prof. UAM dr hab. Waldemar Wołyński
6. mgr Elżbieta Skrzypczak – secretary.
Schedule of the Doctoral School enrolment procedure:
Recruitment fee
The recruitment fee is 75 EUR.
Form of the selection procedure
One stage procedure.
Language of the selection procedure, including interviews:
Polish or English language.
Required documents:
Candidates applying based on exceptional academic achievements must include a scan of the following:
1) For first-cycle graduates: the diploma and supplement.
2) For students completing the third year of a unified master’s programme: a certificate indicating their academic average from years 1 to 3, along with a transcript.
Candidates educated outside Poland must additionally submit:
1) A scanned copy of their higher education diploma and supplement for both first- and second-degree qualifications (Bachelor’s and Master’s), in the original language and a certified translation into English or Polish.
2) A scanned copy of their passport (for foreign applicants).
Evaluation criteria
1. A minimum grade of "very good" or equivalent from a second-cycle (master’s) or unified master’s degree diploma (5 points). Candidates applying on the basis of exceptional scientific achievements (e.g., grant awarded under the “Pearls of Science” competition) will also receive 5 points.
2. Evaluation of up to three documented scientific achievements submitted during registration in the IRK system, assessed according to the detailed criteria of the candidate's accomplishments (maximum of 20 points).
3. Assessment of other documented activities, based on up to three achievements indicated by the candidate, as per the specific criteria for the discipline (maximum of 5 points).
4. The outcome of the interview, which assesses:
a) The candidate’s knowledge and competencies relevant to their intended research, including discipline-specific expertise.
b) Research methodology appropriate to the chosen discipline. The interview has a maximum duration of 30 minutes and awards up to 50 points.
5. Evaluation of the research project prepared for admission, or in the case of recruitment tied to a grant project, the author’s concept for implementing the project (maximum of 20 points). Assessment focuses on:
a) Formulation of the research objective and presentation of the research problem. Page 6 of 9
b) Originality of the research idea and approach to solving the research problem.
c) Methodology suited to the indicated discipline.
d) Awareness of the current state of research, supported by a basic bibliography.
e) The project’s significance for the development of the chosen discipline.
Condition of admission to the Doctoral School
Education Program
Education at the doctoral school lasts 8 semesters. Education at the Doctoral School is offered on the basis of a framework curriculum and an individual research plan and ends with the submission of a doctoral dissertation.
The framework curriculum at the doctoral school includes a catalogue of modules of compulsory classes, elective compulsory classes and optional classes. It also:
1) defines their schedule
2) defines the way in which the classes are taught
3) defines how learning outcomes are verified.
Supervisors
A doctoral student works under the supervision of a supervisor or supervisors, or a supervisor and auxiliary supervisor, who will be appointed by the Deputy Rector at the request of a doctoral student (within 3 months of the start of his or her studies at the Doctoral School).
Scholarships
The scholarship is PLN 5.000 gross-gross for 24 months, than PLN 6.500 gross-gross for next 12 months
The amount should be reduced by 11,26% due to ZUS (social insurance) contributions.

