During the talks at the Innovations in Design Analysis and Dissemination (IDAD): Frontiers in Biostatistics & Data Science in 2022, I took a peek at various website and articles. These were either ones that the speakers mentioned or ones I googled on a topic that they covered. I am placing links to these here in case I want to look at them later.
There is no rhyme or reason to what I decided to include here. This page is not intended to be a comprehensive or representative summary of this meeting.
The fused Kolmogorov filter
Qing Mai, Hui Zou. The fused Kolmogorov filter: A nonparametric model-free screening method. arXiv:1403.7701, and later in the Annals of Applied Statistics. Available in pdf format.
- A methodology to reduce the number of independent variables to a more manageable number.
The SCAD penalty
Kenneth Tay. The SCAD penalty. Statistical Odds and Ends blog, 2018-07-31. Available in html format.
Bayesian designs for extrapolating to pediatric patients
Matthew A. Psioda, Xiaoqiang Xue (2020) A Bayesian adaptive two-stage design for pediatric clinical trials, Journal of Biopharmaceutical Statistics, 30:6, 1091-1108, DOI: 10.1080/10543406.2020.1821704. Article is behind a pay wall
- An alternative to the absolute value penalty used in LASSO regression.
Sample size calculation for survival data without the proportional hazards assumption
Brian Mosier, John Keighley, Milind Phadnis. A Macro to Calculate Sample Size for Studies Using the Proportional Time Assumption. Midwest SAS Users Group 2018 proceedings, Paper HS-083. Available in pdf format.
- While most power calculations use a proportional hazard assumption, it is more intuitive, perhaps, to calculate based on a proportional time assumption. This means, for example, that the time it takes for a certain percentage of the events to occur is twice as long for one group versus the other.