PyMC3 Examples: GLM with Custom Likelihood for Outlier Classification
A worked example of a novel generative model to filter out noisy / erroneous datapoints in a set of observations, compared to alternative methods. Implemented in the probabilistic programming language `pymc3` in a fully reproducible Notebook, open-sourced and submitted to the examples documentation for the PyMC3 project
On contractor day rates
How does an annual salary convert to a day-rate?
Don't call it a comeback
I've been here for years
Survival Analysis: Part1 - What is it?
Survival analysis is long-established within life actuarial work but infrequently used in general data science projects. This series of posts investigates why it's so useful for time-dependent effects, with worked examples.
Tools of the trade (an overview)
We use a variety of software tools for preparing, exploring and modelling data; usually scientific, lightweight and flexible, allowing bespoke insight.
Data science has become a well-established discipline. What is it?
The term 'data science' has been around for several years with many explanations, discussions and breathless over-excitement in the technology and business press. What is it, where did it come from, and who's using it today?