Xiangyuanchai (Cicely) Guo is a second year student pursuing Bachelor of Actuarial Studies in ANU. Early in primary school, her outstanding math performance drew attention from her teachers and parents, who helped provide her with further education on Olympiad Math. It was also her distinguished Olympiad Math grade that gave her access to one of the best high schools in China.
At ANU, while most of her mandatory courses are statistics courses, she gradually developed a strong curiosity about statistics and data analysis. This curiosity encouraged her to use up all her elective courses to study the rest of third year statistics courses, combining her degree with an additional major in statistics.
Cicely has been thinking of having a taste of research for a long time, which is, for her, the natural next step given that she has successfully tackled the coursework and gained a technical base from which to embark on research discovery. In particular, she is interested in statistical modelling. Her research project is on selection effects in binary data, and is supervised by Professor Alan Welsh.
For Cicely the future is unknown yet full of hope, containing infinite possibilities. As a 20-year-old, she is always ready for the opportunities to discover the unknown world. If statistics is the right discipline and research is the right vocation for her, she is willing to take on them as a longer time focus.
Fixing a Mixture Model for Incomplete Data
Informative selection is very difficult to model. Recently, in Section 7.2 of “Maximum Likelihood Estimation for Sample Surveys” (CRC Press), a simple problem with binary data was considered. This project will involve reviewing the presentation in Section 7.2 and doing some simulation work to explore the methods.
The project will involve understanding the importance of taking selection effects into account, understanding the calculations behind maximum likelihood estimation of the parameters of a binary model with selection effects and implementing these calculations in R. Depending on progress, the researcher may do a small simulation.