Skip to main content

Daria Semochkina

Biography

I did my undergrad at Lomonosov Moscow State University; it was a 5-year degree in pure maths. I believe this experience had led to the development of abstract thinking that helps me attack applied problems nowadays. My current PhD project is on Markov Chain Monte Carlo Methods.  These methods are used to sample the data that cannot be sampled directly, so these methods give us with the algorithm that creates a random walk that will eventually provide you with the sample you are looking for.

Research Interests

Calibration of Disease Models for Health Systems Evaluation

 One of the biggest objectives in modern life is to tracks patients as they are diagnosed each year and progress through the disease and treatment course. This project examines calibration of natural history model of HPV virus progression using a fully Bayesian approach.  The model’s predictions of key outcomes depend on a number of unobserved parameters such as transition probabilities between health states. This model facilitates decision making regarding population-level interventions such as screening and vaccination.

The calibration of a natural history model is an inverse problem where the parameters are estimated using data from observed cohorts of individuals at different stages of a disease.  Markov Chain Monte Carlo (MCMC) sampling was used to obtain samples from the distributions for these parameters. This was compared with a simple Monte Carlo approach. A number of different proposals within the MCMC framework were examined to find the best calibration method. Our motivation in this research is to use MCMC methods to explore the distribution of the transition rates, conditional on the information available from the cohort studies.  Some discussion on the use of prior information regarding the physically plausible rates for some of the parameters is also discussed.

The MCMC approach is shown to be working better than the simple Monte Carlo independent sampling. However, within the MCMC framework different proposals showed different convergence and mixing properties. Finding the balance between changing proposals to achieve better results has proven to be a great challenge.

Teaching Interests

I teach tutorials for 
MS4212 "INTRODUCTORY DATA ANALYSIS"​,
MS4217 "STOCHASTIC PROCESSES"​,
MS4214 "STATISTICAL INFERENCE"​
MS4222 "INTRODUCTION TO PROBABILITY AND STATISTICS"
MS4131 "LINEAR ALGEBRA 1"
MA4704 "TECHNOLOGICAL MATHEMATICS 4" 

Professional Activities

Award

  • 2014 - Best poster

Employment

  • 2013 Moscow State University - Project manager
  • 2011 Star Manager Consulting - Business Analyst
  • 2010 Energy Consulting - Assistant Consultant

Education

  • 2004 Lomonosov Moscow State University - MSc

Committee

  • Athena Swan,
    ECU’s Athena SWAN Charter was established in 2005 to encourage and recognise commitment to advancing the careers of women in science, technology, engineering, maths and medicine (STEMM) employment in higher education and research. 
    Our department applied and received Athena Swan bronze award in December 2015.
  • Women in Maths Day, Organising  the Women in Maths Day

Language

  • English
  • Russian
  • French