Posts

The MaxCombo Test severely inflates the type 1 error rate

“The MaxCombo Test severely inflates the type 1 error rate”, by Carl-Fredrik Burman and myself has just been published in JAMA …

{nphRCT} on CRAN

I’m delighted that our new R package {nphRCT} is available on CRAN. Special thanks to Isobel Barrot. Her R skills have made this …

Pseudo-observations 2

This is a quick follow up to my previous post on pseudo-observations where I reproduced Table 1 from “Causal inference in survival …

Pseudo observations on pseudo-observations

Idea / motivation I’ve set myself the challenge of reproducing the results in Table I of “Causal inference in survival analysis using …

Conditional and unconditional treatment effects in RCTs

Small simulation study Quick function to simulate data from an RCT with equal allocation to treatments 0 and 1, a single covariate, and …

What's the difference between fixed, random, independent, type I, type II, (non)prognostic, (non)ignorable, and (non)informative censoring?

I already feel exhausted after writing the title. I’m not really going to explain all these differences, sorry. I’m not capable. And …

Be careful with standard errors in `survival::survfit`

(Mild) panic In my previous post I looked into how survival::survfit produces standard errors and confidence intervals for a survival …

Confidence interval for a survival curve based on a Cox model

A colleague caught me out recently when they asked about a confidence interval for a survival curve based on a Cox model. This can be …

Notes on concordance

Motivation I don’t understand the details of concordance in survival analysis, so I’m trying to explain it to myself. Probabilistic …

Trouble with tau

This post is to express some minor frustration with some papers I’ve read recently evaluating the performance of restricted mean …