Rstats

Direction of Bias From Nondifferential Misclassification

In this post, I simulate 3 scenarios of non-differential classifications of exposure as demonstrated in Yland et al., 2022: Specificity of exposure 90% and sensitivity 60% Specificity of exposure 90% and sensitivity 70% Specificity of exposure 90% and sensitivity 80% Each of these scenarios will be simulated in 10 thousand datasets of varying sizes: n=100, n=1000, n=10000.

Competing events survival analyses

This is an account of R code by Paloma Rojas-Saunero, PhD available here, which I “translated” to tidyverse code and added some clarifications and figures. This is a longer read, but, hopefully, also is easier to follow for those who use tidyverse.

Chp7 Splines - Chp 8 KNN

Boook club "Hands-On Machine Learning with R"

Using `WeightIt` R package for causal inference analyses

I recently discovered WeightIt R package and was very happy with its functionality and performance. I “delegated” my code computing IPTW to WeightIt and it was faster while producing the same results, as expected.

Antibiotics utilization in Denmark: using R to solve practical tasks in epidemiology

Intro This is a blog post-chaperon for the R-Ladies meet-up Abuja. We are talking R in epidemiology, practical aspects of being R-user and going through data-supported example of how to use R for data aggregation and visualization.

Income inequality: OECD data

Data In this post I explore income inequality. The data comes from OECD, where inequality is defined as household disposable income per year. Main income inequality markers I use from the dataset are:

TidyTuesday 2022: week 1

Data I’m going to use the data I’m intimately familiar with: medication utilization in Denmark. I will visualize antidepressants use patterns. I’ll use palettes from my {hermitage} package. My favourite palettes so far are madonna_litta and hermitage_1.

Kaggle ML survey 2021

I am a doctoral student. I often wonder what my future holds in this brave and largely liberated according to some world. In EU, although 48% of doctoral graduates were women according to She Figures 2021, only 34% of researchers are women and only 24% of heads of higher education institutions are women.

TidyTuesday Starbucks Data

Data The data I use are available here. Let’s go ✌ I have no initial idea what I want to present and so made several exploratory plots to see what I deal with.

TidyTuesday Spice Girls Data

Data I use data by Jacquie Tran available here. Let’s go ✌ I chose to plot the audio features of Spice Girls tracks: danceability, energy, speechiness, acousticness, valence, liveness, and instrumentalness.