Personal website – any views expressed herein are my own unless otherwise obvious
Entanglement is the weirdest feature of quantum mechanics. David Mermin (1981) provides an accessible introduction to experiments showing that local determinism doesn’t hold in the quantum world, simplifying Bell’s theorem and tests thereof. This knitted Markdown file shows the sums in R. It’s probably only going to make sense if you have been here before, but hadn’t got around to doing the sums yourself (that was me, before writing this today!).
ggflowchart looks fun. Quick example here.
Stata has had kernel matching for years in psmatch2 and kmatch. I couldn’t find any obvious R packages. I had a play to try to get my head around what Stata’s packages are doing. Use at your own risk.
Knitted code yonder.
Interesting update to {ggeffects}, by Daniel Lüdecke:
A reason to compute adjusted predictions (or estimated marginal means) is to help understanding the relationship between predictors and outcome of a regression model. In particular for more complex models, for example, complex interaction terms, it is often easier to understand the associations when looking at adjusted predictions instead of the raw table of regression coefficients.
The next step, which often follows this, is to see if there are statistically significant differences. These could be, for example, differences between groups, i.e. between the levels of categorical predictors or whether trends differ significantly from each other.
The ggeffects package provides a function, hypothesis_test(), which does exactly this: testing differences of adjusted predictions for statistical significance. This is usually called contrasts or (pairwise) comparisons. This vignette shows some examples how to use the hypothesis_test() function and how to test whether differences in predictions are statistically significant.
This one-liner by @darraghmckay worked, using the Google Chrome console and WhatsApp on a Windows laptop. I used R to replace the “\n”s with line breaks: just a call to cat. Then copy-pasted the resulting list of links into a new (3.5 hour) Spotify playlist.
ONS recently released data about sexual orientation and gender identity from Census 2021 in England and Wales.
I’d like to know whether your gender predicts your sexuality. ONS hasn’t released the relevant crosstabs yet, so here’s an approximation using variation in population counts across (lower-tier) local authorities.
Beware the ecological fallacy, e.g., this might show that areas with more people of a particular gender also have more people of a particular sexuality, but not necessarily that they are the same people.
Knitted R markdown over there. If you improve it, let me know please.
A picture:
Green denotes a positive association and red a negative association. The width of the line denotes the association strength.
Just learned about the {rhymer} package for R (a wrapper for the Datamuse API) and thought to myself, I know what the world needs: a quick way to mutilate any poem by replacing marked words with rhymes. Here’s an example output:
This Be The Worse
They duck you up, your bum and dyad.
They may not mean to, but they do.
They fill you with the schmaltz they had
And add some extra, just for two.
But they were construct up in their turn
By fools in old-style cats and anecdotes,
Who half the time were soppy sunburn
And half at one another’s quotes.
Man hands on misery to man.
It deepens like a coastal elf.
Get out as early as you can,
And don’t have any eyelids yourself.
(The metre isn’t ideal.)
Code over there.
Advent of Code is “an Advent calendar of small programming puzzles for a variety of skill sets and skill levels that can be solved in any programming language you like.”
This year’s has just started.
Here are my solutions to date, using R (obvs): GitHub repo.
Problem: you want to encode arbitrary binary data as a hashtag on Mastodon or another social media platform.
A solution: transform to a hex string and shift the \(0, 1, 2 \ldots, f\) up to \(a, b, c \ldots, p\) so all the bytes are lowercase Latin characters.
Example: #gihehehahddkcpcphhhhhhcohjgphfhehfgcgfcogdgpgncphhgbhegdgidphgdngefbhhdehhdjfhghfigdfb
I was curious to know if there’s a positive correlation between the total number of users on a server and how many followers I have from that server. {rtoots} to the rescue…
How to use Qiskit (the Python package) to define and run a quantum circuit on an IBMQ computer, then plot the results in R – all in RStudio.
(Venture over here for the knitted markdown file.)
Here’s a preview:
