Rathish
Author – Ritesh Ranjan: TU Delft has announced a four-year PhD position in Mathematical Statistics for candidates interested in advanced statistical theory, dependence modelling and copula-based methods. The position is offered by the Department of Applied Mathematics within the Faculty of Electrical Engineering, Mathematics and Computer Science, commonly known as EEMCS.
This opportunity is designed for highly motivated MSc graduates who want to work on mathematical statistics at a deep theoretical level while also contributing to methods that can be applied in real-world domains. The research will focus mainly on copula theory, conditional copulas, information geometry and Pair-Copula Bayesian Networks.
For students aiming to build an academic or research-focused career in statistics, probability, mathematical modelling or data science, this PhD position offers a strong platform at one of \
About the PhD Position
The PhD role at TU Delft is a full-time, four-year research appointment. The selected candidate will work on new theory and methodology for dependence models, especially statistical models based on copulas.
Copulas are powerful mathematical tools used to describe dependence structures between random variables. In simple terms, they help researchers understand how different variables move together, even when those variables have different distributions. This makes copulas especially useful in fields such as finance, risk management, insurance, hydrology, economics and health sciences.
The project is expected to combine rigorous mathematical work with computational implementation. Candidates with a strong foundation in mathematical statistics and good programming skills will be well suited for this role.
Key Research Areas
The PhD project is divided into three major research directions. Each area focuses on a different aspect of dependence modelling and statistical theory.
1. Conditional Copulas and Conditional Dependence Models
The first part of the project focuses on conditional copulas and conditional dependence models. These models study how the relationship between two or more variables may change depending on another random variable or vector.
For example, in finance, the dependence between two assets may be influenced by a third asset, a market factor or an economic condition. Conditional copulas help researchers model such changing relationships more accurately.
This research direction is important because real-world systems rarely have fixed dependence structures. Markets, weather systems, health outcomes and social phenomena often change depending on external conditions. Developing better conditional dependence models can therefore improve statistical analysis and prediction in many applied areas.
2. Information Geometry for Copulas
The second research direction aims to build a theory of information geometry for copulas. Information geometry is an interdisciplinary field that combines probability, statistics and geometry.
It studies statistical models as geometric spaces. This allows researchers to examine concepts such as distance, direction, curvature and angles within families of probability distributions. Applying these ideas to copulas could lead to new ways of understanding and comparing dependence structures.
This is a highly theoretical part of the project and may appeal to candidates who enjoy abstract mathematical thinking. At the same time, the results could contribute to practical statistical tools for model selection, estimation and inference.
3. Pair-Copula Bayesian Networks
The third direction focuses on integrating discrete random variables into Pair-Copula Bayesian Networks, also known as PCBNs. These models combine the flexibility of copula-based modelling with the structure of Bayesian networks.
Traditional Gaussian Bayesian networks can be limited because they often rely on restrictive distributional assumptions. Pair-Copula Bayesian Networks are more flexible because they break complex multivariate distributions into smaller bivariate components while allowing arbitrary margins.
The project will also involve developing structure-learning algorithms. These algorithms help identify the relationships between variables in a network. This research area is particularly relevant for modern data analysis, where complex systems often involve many interacting variables.
Who Can Apply?
TU Delft is looking for candidates who either already hold or are close to completing an MSc degree in Mathematical Statistics or an equivalent field. The position is suitable for applicants with strong mathematical ability, research curiosity and the motivation to develop new models and theories.
Applicants should have strong programming skills in R. Knowledge of C++ will be considered an advantage. Since the work involves both theory and computation, candidates should be comfortable with mathematical reasoning as well as implementation.
Fluency in English is also essential. PhD candidates at TU Delft are expected to participate in doctoral education courses, communicate with international colleagues, write academic papers and complete a doctoral thesis in English.
Why Copula Theory Matters
Copula theory has become an important area in modern statistics because it allows researchers to model dependence separately from marginal distributions. This is useful when variables are connected in complex, non-linear or asymmetric ways.
In finance, copulas can help model the dependence between assets, markets or risk factors. In insurance, they can be used to understand the joint behaviour of different types of claims. In hydrology, they can help study relationships between rainfall, river flow and flood risk. In health research, copulas can support the modelling of related outcomes or risk indicators.
Because many real-world problems involve dependence between several variables, better copula models can lead to more accurate analysis and decision-making.
About TU Delft
TU Delft is one of Europe’s most respected technical universities. It is known for combining science, engineering and design to address major global challenges in areas such as energy, climate, mobility, health and digital society.
The university promotes an international and inclusive research culture. For PhD candidates, this means access to a dynamic academic environment, expert supervision and opportunities to collaborate with researchers from different backgrounds.
The Faculty of Electrical Engineering, Mathematics and Computer Science is one of TU Delft’s major faculties. It has more than 1,000 employees and around 4,000 students. Research within the faculty covers a wide range of fields, including applied mathematics, artificial intelligence, future chips, sustainable electricity grids, software systems and data science.
Appointment Structure and Salary
The PhD appointment is structured as a four-year position divided into two contracts. The first contract will run for 1.5 years. After this period, the candidate will undergo an official go/no-go assessment. If progress is satisfactory, the appointment will be extended with a second contract of 2.5 years.
The salary follows the Dutch university collective labour agreement. It starts at €3,059 gross per month in the first year and increases to €3,881 gross per month by the fourth year. This is based on a full-time 38-hour working week.
The benefits package also includes 8% holiday allowance and an 8.3% end-of-year bonus. Additional benefits may include flexible work schedules, health insurance discounts and a monthly work costs contribution.
Graduate School and Research Support
Selected PhD candidates will be enrolled in the TU Delft Graduate School. The Graduate School provides a structured doctoral education programme designed to support both academic and professional development.
Candidates receive guidance in disciplinary research skills as well as transferable skills such as scientific writing, communication, project management and career planning. This support helps PhD researchers grow not only as subject experts but also as independent scholars.
The combination of expert supervision, a strong research community and international exposure makes TU Delft an attractive destination for doctoral study.
Application Process
Interested candidates must apply through the official TU Delft application portal by 14 June 2026.
Applicants are required to submit the following documents:
- CV
- Motivational letter
- Grades of BSc and MSc courses
Candidates who are still completing their MSc may submit the grades currently available.
For further information about the position, applicants may contact Dr. Alexis Derumigny at a.f.f.derumigny@tudelft.nl.
Why This PhD Opportunity Stands Out
This PhD position stands out because it combines strong mathematical theory with relevant applications in modern statistics. The selected candidate will work on advanced dependence modelling, a field that continues to grow in importance across finance, risk analysis, data science and applied research.
The role is especially suitable for students who enjoy mathematical depth, independent research and the challenge of developing new statistical tools. With TU Delft’s global reputation, strong faculty environment and structured doctoral support, this opportunity can be a major step toward a high-level research career.
For ambitious MSc graduates in mathematical statistics, this TU Delft PhD position offers a chance to contribute to cutting-edge research while working in one of Europe’s leading academic institutions.
FAQs
1. What is the main focus of the TU Delft PhD position in Mathematical Statistics?
The main focus is on dependence modelling using copula-based statistical models. The project includes conditional copulas, information geometry for copulas and Pair-Copula Bayesian Networks.
2. Who is eligible to apply for this PhD position?
Applicants should hold or soon complete an MSc degree in Mathematical Statistics or an equivalent field. Strong mathematical skills, R programming ability and fluency in English are required.
3. What programming skills are required for the position?
Strong programming ability in R is required. Knowledge of C++ is considered an advantage but is not listed as mandatory.
4. What is the salary for the TU Delft PhD position?
The salary starts at €3,059 gross per month in the first year and rises to €3,881 gross per month by the fourth year, based on a full-time 38-hour week.
5. What is the application deadline?
The application deadline is 14 June 2026. Candidates must apply through the official TU Delft application portal.