Two widely used employment indicators, ADP National Employment Report (ADP) and Non-Farm Payrolls (NFP), provide valuable insights but are controversial. Some analysts claim that there’s no correlation between ADP and NFP numbers, but a deeper analysis shows there’s more nuance beneath the surface. …

Healthcare intertwines biological causal effects and systematized interventions. As a result, a statistical paradox commonly emerges when analyzing real-world evidence: harmful risk factors may be associated with better outcomes. This paradox is especially prevalent with systematized procedures which follow standardized protocols. For example, admission policies play a crucial role in…

Investing genius Warren Buffett is renowned for his successful investment strategy and the remarkable returns he has generated. However, one aspect of his approach that often raises eyebrows is his reluctance to invest in tech companies. Buffett claims he “doesn’t understand tech companies” and prefers to leave them in the…

The inflation/deflation debate is confusing. The fundamental challenge is that economic “inflation” is composed of multiple factors and different analysts like to focus on different factors without specifying their assumptions. At a high level, we can model inflation as a change in the product of money velocity and money supply…

For many sports, the roar of the crowd can provide an extra boost of energy for players and create an electrifying atmosphere for fans. But what’s the impact of that atmosphere on the gameplay? …

#ChatGPT is amazing! Here’s a fun little example I made where it goes wrong, prioritizing pattern matching over logical reasoning. Inspired by this explanation of psychological biases underlying Monty Hall intuition, I decided to ask #ChatGPT. I begin playing a version as the host. …

Ben Lengerich Which hospitalized patients are the most likely to die? In our new paper, we show that the patients at highest risk of death are often not the most unhealthy patients overall, but rather the most unhealthy of the patients who don’t get aggressive treatment. For example, serum creatinine…

This blog post describes a model we built to understand and visualize the implications of limited diagnostic testing capacity on the apparent spread of SARS-CoV-2 in 2020.

Ben Lengerich Why Sample-Specific Models? Most ML systems (including modern deep learning) assume that one single model can summarize complex processes, as long as the model has large enough representational power. …