Understanding the causal relationships between variables is crucial for making informed decisions and predictions. The Python package causal-learn offers a powerful toolkit for causal discovery, enabling researchers and practitioners to uncover these relationships from observational data. In this blog post, we’ll explore how to leverage causal-learn to identify and analyze causal structures, providing a step-by-step guide to get you started on your causal discovery journey.

In the rapidly evolving landscape of app development, managing and persisting data efficiently is crucial. Whether you are building a photo gallery app, a note-taking tool with image attachments, or any application that involves handling images, the way you store and retrieve those images can significantly impact performance and user experience.

SwiftUI, Apple’s innovative UI toolkit, has transformed the way developers create user interfaces for iOS, macOS, and beyond. One of the fundamental tasks in app development is handling data, and JSON (JavaScript Object Notation) is a widely-used format for data interchange. In this blog post, we’ll delve into the essentials of loading JSON data in SwiftUI. We’ll start by exploring how to load JSON data directly as a String—useful for quick tests or when dealing with small data payloads. Then, we’ll advance to loading JSON from a file, which is a common scenario for apps that consume complex data structures or large datasets.

In every statistical analysis, the selection of covariates in regression models is akin to navigating a labyrinth. One common heuristic advises controlling for as many variables as possible, under the presumption that it yields more conservative tests of hypotheses. However, this approach is not without pitfalls. Adding wrong or extraneous variables can distort results, introducing spurious effects and even reversing the signs of actual causal relationships. In the following, I will provide an example of how this can happen.

After reading Rasmus Bååth’s blog Why pandas feels clunky when coming from R, I decided to practice more polars and translate his dplyr examples to polars.

In the realm of healthcare, education, and policy-making, understanding the effects of treatments and interventions is paramount. However, discerning causality from mere correlation can be challenging. In this blog post, I invite you on a journey to unravel the complexities of estimating the true impact of interventions. From exploring randomized controlled trials to delving into the intricacies of confounding variables, join me as we navigate through the fascinating world of causal inference and uncover the methods and insights essential for evidence-based decision-making.