stephematician

👋 My name is stephematician. I'm a mathematician, programmer, and former researcher in lung cancer screening and tobacco control. Have a look around at my work in R, python 🐍, and C++.

Highlights

• linuxgreenscreen: A virtual 'green screen' 🟩 for GNU/Linux, see this announcement.
• literanger: Random forest training and prediction. Available on CRAN.
• Gitlab CI/CD components for R.

I'll soon be back on twitch.tv/stephematishun to chat, play retro games (like System Shock [1994]), and maybe code.

Publications

Full list: Google Scholar 🔗.

• 2023. Fifty-year forecasts of daily smoking prevalence: can Australia reach 5% by 2030?. Tobacco Control. Published online 22-05-2023. doi:10.1136/tc-2022-057624.
• 2022. Bayesian calibration of simulation models: a tutorial and an Australian smoking behaviour model. arXiv:2202.02923.
• 2018. Estimating the cost-effectiveness of lung cancer screening with low-dose computed tomography for high-risk smokers in Australia. Journal of Thoracic Oncology, 13(8), 1094-1105. doi:10.1016/j.jtho.2018.04.006.
• 2014. On the free-surface flow of very steep forced solitary waves. Journal of fluid mechanics, 739, 1-21. doi:10.1017/jfm.2013.590.

Other projects

• pclm: Ungroup 'binned' data using penalised composite link models in R.
• effectspline: Use splines in age-period-cohort models in R.
• Private projects (on request) for R:
  • mutual information of mixed-type data using approximate nearest neighbours, and;
  • multiple imputation via random forests.