In this post we use statistics and historic weather data to help answer the question “is it really getting warming in December in NJ?” Using progressively more complex linear models we find some surprising results about what’s happening to weather in the area.

In this post we explore the very curious logit-normal distribution, which appears frequently in statistics and yet has no analytical solutions to any of its moments. We end with a discussion of Nietzsche and what Dionysian statistics might mean. It’s a bit of an odd post that’s for sure.

Exploring how Frequentist and Bayesian methods differ when looking at the common problem of predicting misfires in extra-dimensional alien technologies. We explore the nice property that Bayesian statistics are mathematically correct and then consider that sometimes being “technical wrong” does work out.

In part 2 of our series exploring the relationship between Inference and Prediction we explore modeling Click Through Rate as a Statistical modeling problem so that we can more deeply understand the role of Inference in modeling.

This post is the first in a three part series covering the difference between prediction and inference in modeling data. Through this process we will also explore the differences between Machine Learning and Statistics. We start here with statistics, ultimately working towards a synthesis of these two approaches to modeling

Survivorship bias is a very common topic when discussing statistics and data science, but identifying survivorship bias is much easier than modeling it! In this post we take a look at modeling survivorship bias in MLS house listing using mathematics, Python and JAX and see just how challenging of a task it is.

Thompson sampling is a technique to solve multi-armed bandit problems but choosing actions based on the probability they offer the highest expected reward. In this post we learn about this technique by implementing an ad auction!

Bayesian statistics rely heavily on Monte-Carlo methods. A common question that arises is “isn’t there an easier, analytical solution?” This post explores a bit more why this is by breaking down the analysis of a Bayesian A/B test and showing how tricky the analytical path is and exploring more of the mathematical logic of even trivial MC methods.