Columbia University

What you'll study

The PhD curriculum combines rigorous coursework in mathematical foundations with intensive research training. Early-stage students complete core courses in optimisation (convex and nonconvex methods), probability and stochastic processes, statistical inference, and numerical methods. Typical modules and topics include:

• Convex analysis and optimisation algorithms
• Stochastic processes, queueing theory and applied probability
• Statistical learning, inference and high-dimensional statistics
• Stochastic optimisation and control
• Simulation methods and Monte Carlo techniques
• Network models, logistics and supply chain optimisation
• Computational methods for large-scale optimisation and data-driven decision-making

Students also take elective courses across related departments (for example computer science, statistics, industrial engineering, and applied mathematics) to tailor their programme to a chosen research area. The degree emphasizes passing qualifying or breadth examinations, completing advanced seminars, forming a dissertation committee, and producing a PhD dissertation based on original research under faculty supervision.

Entry requirements

Applicants are expected to have a strong quantitative background. Typical successful candidates hold a bachelor’s or master’s degree in engineering, mathematics, statistics, computer science or a related field with substantial coursework in linear algebra, real analysis, probability, and optimisation. The application usually requires:

• Academic transcripts demonstrating strong grades in quantitative subjects
• A curriculum vitae outlining research experience, publications or relevant projects
• Letters of recommendation that speak to research potential and mathematical maturity
• A statement of purpose describing research interests and fit with the department
• Proof of English language proficiency for applicants whose first language is not English

Admissions committees look for demonstrated ability to undertake independent research. Some applicants have prior research or industry experience, and many successful applicants have completed master's-level coursework or research projects. Applicants should consult the department’s admissions pages for any additional requirements or optional test policies.

Career prospects

Graduates of the PhD programme pursue a wide range of careers that leverage deep quantitative and modelling skills. Common pathways include:

• Academic careers as postdoctoral researchers and tenure-track faculty in operations research, industrial engineering, statistics or computer science
• Research scientist roles in major technology companies, AI and machine-learning groups, and high-performance computing labs
• Quantitative research and modelling positions in finance, risk management and trading firms
• Analytics and optimisation roles in logistics, supply chain, transportation and energy sectors
• Consulting and policy roles that require advanced decision analytics and stochastic modelling

Alumni commonly work in environments that demand a combination of theoretical insight and practical implementation skills, from building production-scale optimisation systems to publishing in leading academic journals.

Why study at Columbia University

Columbia’s programme is embedded in a department with a long-standing strength in both theoretical operations research and applications. Students benefit from access to a diverse faculty whose research spans optimisation, stochastic systems, statistics and machine learning. Being in New York City offers proximity to banks, technology firms, consultancies and start-ups for industry collaboration and internship opportunities.

Additional advantages include interdisciplinary research centres, active seminar and colloquium series, and access to university-wide resources such as high-performance computing and data-science initiatives. The environment encourages collaboration across departments and with external partners, supporting both fundamental research and practical impact.