Historical records of population heights provide information about the socio-economic trends prevalent at the time. However, the most commonly available historical records are not usually representative of the population at large. For example, the army preferred to recruit the taller and healthier men while the navy also had height restriction due to the cramped condition on the ships. The truncation was not precise but rather stochastic. Using a stochastically truncated sample to estimate the underlying population distribution is a common topic in socio-economical historical studies of human stature. However, the methods are often stepwise, inelegant and biased. Finding an elegant statistically sound way to use a doubly truncated sample to fit the underlying distribution will be the topic of this project.

Key qualifications and skills

Bayesian statistics, MCMC algorithms, good programming skills (R preferred)

Does the project come with funding

No

Final date for receiving applications

Ongoing

Keywords

Bayesian statistics