13127 - Postdoctoral Research Associate in Statistical Computing

University of Edinburgh; City of Edinburgh; On-site; GBP 41,000 - 49,000; Full time

Job summary

A leading university in the UK is seeking a postdoctoral researcher in statistical computing methods. This position focuses on developing scalable smooth statistical modelling methods and implementing them in a state-of-the-art Python package. Candidates should hold a PhD in statistical computing or a related field and have strong skills in mathematical and software engineering as well as a solid publication record.

Qualifications

• A PhD in statistical computing methods or closely related field.
• Proven statistical publication track record, with a sound methods component.
• Strong mathematical, numerical and software engineering skills.
• Good knowledge of modern statistical regression, particularly smoothing and mixed models.
• A strong background in Bayesian and frequentist statistical methods.

Responsibilities

• Develop statistical computing methods for general smooth regression models in large data settings.
• Investigate and implement smoothing parameter and uncertainty quantification methods.
• Develop a native Python package for smooth regression modelling.
• Collaborate with industrial partners, Simon Wood, and project students.

Skills

Statistical computing methods

Software engineering

Smoothing and mixed models

Bayesian methods

Frequentist methods

Education

PhD in statistical computing or related field

Tools

Python

Job description

Overview

Grade UE07 £41,064 – £48,822 per annum
CSE / School of Mathematics
Full time: 35 hours per week
Fixed term: from 1st January 2026 until 31st December 2029

The Opportunity

We are looking for a postdoctoral researcher in statistical computing methods, to work on the next generation of scalable and stable smooth statistical modelling methods, and their implementation in a state of the art python package. This 4 year project is collaborative with industrial partners and Simon Wood (the author of R package mgcv for smooth regression modelling).

The researcher will work with Simon Wood, industrial partners and students on developing statistical computing methods for general smooth regression models in large data – large model settings, and on innovative smoothing parameter and uncertainty quantification methods for general smooth models that approach O(np) computational complexity. As well as conventional academic methods development, a second strand of the work aims to produce the state of the art native python package for general smooth regression modelling, improving on mgcv, incorporating the new methods and providing seamless integration with industrial python based data analysis workflows.

Responsibilities

• Develop statistical computing methods for general smooth regression models in large data settings.
• Investigate and implement smoothing parameter and uncertainty quantification methods for smooth models with near O(np) computational complexity.
• Develop a native Python package for general smooth regression modelling, improving on mgcv and integrating with industrial Python workflows.
• Collaborate with industrial partners, Simon Wood, and project students.

Qualifications

• A PhD in statistical computing methods or closely related field.
• Proven statistical publication track record, with a sound methods component.
• Strong mathematical, numerical and software engineering skills.
• Good knowledge of modern statistical regression, particularly smoothing and mixed models.
• A strong background in Bayesian and frequentist statistical methods.