A running list of some of my favorite online statistics resources - including blog posts, visualizations, interactive simulations, and text books. Blog posts & cheat sheets Common statistical tests are linear models: A really clear explanation of how many common statistical tests can be conceptualized as types of linear regressions. An Introduction to Hierarchical Modeling:… Continue reading Favorite statistics resources
Interaction analyses – Appropriately adjusting for control variables (part 4)
In an interaction analysis, the probability of a false-positive result increases as the correlation between our covariate and predictor increases, and as the effect of our covariate x predictor interaction increases. The extremely simple solution is to include all covariate-by-predictor interactions in your model!
Researchers want you to tell them how the pandemic has affected your mental health
A quick intro to block permutations and bootstraps for analyzing hierarchical data
Clustered data - such as multiple observations per individual, animal, or cell - are quite common in neuroscience research. Here I walk through an introduction to one approach to adjusting your analyses for clustering - block permutations and bootstrapping - that is widely applicable and makes very few assumptions.
Correlates of rock climbing ability
Here is my analysis of 'what correlates with how well someone climbs?'.
Using the new Turbo palette from Google in R
Interaction analyses – How large a sample do I need? (part 3)
Interaction analyses – Interpreting effect sizes (part 2)
Interaction analyses – Power (part 1)
In this series I try to convey a couple of insights about power for interactions in linear regressions. First, how to do a power analysis for a interaction in a linear regression (this post), then interpreting the effect size of a interaction (part 2), and finally thinking about how large (or small) an effect size it is reasonable to plan for (part 3).
Elizabeth U Baranger
Elizabeth Urey Baranger (born Gertrude Elizabeth Urey) (09/18/1927 - 05/30/2019) was an American theoretical physicist and senior administrator at the University of Pittsburgh.
The two-body problem: postdoc edition
We managed to secure postdoc positions in the same city, and were able to move there together! Here are a handful of things we did that we think helped us pull this off.
Improving Genetic Prediction: Data cleaning & Meta-analysis
In my first post I showed how to build a simple model, using only genetic information, that predicts height, with 53% accuracy in an independent sample. In this post I’m going to improve that model, which will ultimately result in a model with 64% accuracy in an independent sample - an 11% improvement!
Showing your data; Scatter-box-violin plots
Some advice on getting a post-baccalaureate research assistant position
I've heard it said that if you send the same email more than a couple of times, you might want to turn it into a blog post. So here is some advice I've given to several people over the past few years, on how to get a job as a research assistant (RA) after college.
On Predicting Traits with Genetics
crowdAI is a VERY cool site that hosts machine learning competitions. They recently hosted a competition using data from openSNP, a website that lets anyone upload their genetic data, making it publicly available. About 3,500 people have uploaded some amount of their genetic data (many people just upload the mitochondrial portion), and 921 also reported their height. The crowdAI competition was, very simply, can you predict Height in a subset of these people, using the genetic data from everyone else to train a model?
And? I won!!! My best model predicts 53.45% of variance in Height, while the current next-best predicts 48.62%. But there’s a catch, even though it was a “machine learning”" competition, I didn’t actually use any ML…oops
This post is intended to give an explanation of why I participated and how I won, with the hope that others with more expertise in ML can benefit from my knowledge in genetics....