Record Statistics of Non i.i.d Random Variables

My research project will explore the behaviour of record statistics in a non-i.i.d. setting with a focus on sequences with linear drift such. The expected value of the current record after n many time steps will be calculated as well as determining the distribution of the waiting time between records. This will use the integral formula for the probability of a record at time n. These results will be explored through the use of simulations utilising the generalised Pareto distribution because of how its maximal values can fall into any of the 3 extreme value distributions established under the Fisher–Tippett–Gnedenko theorem.

Angus Stewart

Macquarie University

Angus is studying a Bachelor of Actuarial Studies and Bachelor of Science majoring in mathematics with a minor in statistics at Macquarie University. He took his first steps into statistical research last summer culminating in a presentation at the Australasian Council of Undergraduate Research and is excited to complete further research this summer about record statistics and extreme value theory extending the concepts introduced during his studies.

He is also an active member of the mathematical community at university through participation in the annual ASA Datafest which his team has won twice, the Simon Marais Mathematics Competition and as president of the Mathematics and Statistics Society following its recent resurrection.

You may be interested in

Runqiu Fei

Runqiu Fei

Predicting Stock Price Volatility within a Month – A Functional Data Analysis Approach
Kyan Percevault

Kyan Percevault

Deep Learning in Hidden Markov Models
Hugo Fellows-Smith

Hugo Fellows-Smith

Regularity of solutions to elliptic boundary value problems near cusp domain boundaries
Edward Mirco

Edward Mirco

Solving Graph Colouring Problems via Convex Optimization
Contact Us

We're not around right now. But you can send us an email and we'll get back to you, asap.

Not readable? Change text.