For its accuracy and its communicability, the logistic regression model is one the most frequently used methods in the field of predctive modeling for models in which the dependent variable is binary. Results from logistic regressions are often reported “the academic way”, i.e., in the form of estimates and its standard error. While estimates indeed are interesting, what’s perhaps even more interesting is the accuracy of the model. In this post I will provide you with a number of visualizations we can do to to illustrate this. [Read More]

Systembolaget, the Swedish alcohol retailing monopoly, generously offers semi-detailed stock data through an open API. The data reveals in which stores specific products currently are in stock, but not the actual stock number. In this document, I will show you how to access the data using R and how to cluster stores based on this. A summary of the data is given by Systembolaget in Swedish here: https://www.systembolaget.se/api/. Data The data consists of three XML trees; one for product data, one for store data, and one for stock data. [Read More]

Here we go.