@bolkerb@cjackstats I didn't know about :: to call a function before writing my 1st #rstats 📦🤔 @bolkerb means that there is also a function called deltamethod implementing the #DeltaMethod in his package 📦 emdbook which comes w/ his great 📖 'Ecological Models and Data'
2 nice papers about the #DeltaMethod: ‘Who Invented the Delta Method?’ by Jay M. Ver Hoef and ‘Approximating variance of demographic parameters using the delta method: A reference for avian biologists’ #OpenAccess by @LarkinPowell
A relative risk (φ), for example, is a ratio of probabilities p1/p2. However, estimating its variance is really difficult! The #DeltaMethod allows us to approximate the risk with the derivative (green line) and then estimate an easier variance using this approximation.
I always forgot about it, so I wrote a short post on how to implement the #DeltaMethod in #rstats - given an estimate and its standard error (SE), how to calculate the SE of any function of this estimate w/ awesome #msm package by @cjackstats
A common way to build confidence intervals is the #DeltaMethod. It generates CI of numbers, vectors, and functions.
We've made a tutorial showing how the functional delta method works and its connection with the influence function in #epi#tutorial
🧵
The variance of the #influencefunction usually is a good approximation to the variance of your estimator and you can easily calculate the asymptotic confience intervals in #rstats with our examples: