5 artifacts
A step-by-step walkthrough of how a decision tree picks where to split a node. With five observations and four candidate thresholds, the artifact traces the RSS computation for each split and lets the reader see why greedy search gains in clarity what it loses in global optimality. Interactive animation and live computation, bilingual.
An interactive scrollytelling artifact explaining the bias-variance tradeoff through the U-shaped test-error curve from ISLR Chapter 2. Six visual steps: from raw data to cross-validation, rendered in a sumi-e ink-on-washi aesthetic. Bilingual: Spanish and English.
What do an actuary and a volcanologist have in common? Both try to predict rare events with enormous consequences. This note reviews how eruption intervals for Popocatépetl and Galeras are modeled with log-normal distributions, Markov models, and renewal processes. The open questions at the end are genuinely fascinating. Personal applied probability note.
Same question, two ways of thinking: the classical approach says there's not enough evidence, the Bayesian one says B is almost certainly better. This summary explores what happens when you stop relying solely on the p-value and start thinking in terms of expected utility to decide. Personal statistical exploration project. Available in both languages.
Five years of Delhi temperature told through data: finding the seasonal pattern hidden in the noise, separating real trend from random variation, and building a model in R that can anticipate what comes next. The kind of exercise where statistics stops being abstract and becomes weather. Course project for Survival Analysis and Time Series, UNAM.