Data Science Tutorials

Learn what minima and maxima are, how local and global extrema differ, and how to find them using critical points, first and second derivative tests, and the closed-interval method.

Learn what an invertible matrix is, how to determine if a matrix is invertible, and why matrix inverses matter in linear algebra and data science.

Learn how SARIMA extends ARIMA to handle seasonality, understand its seven parameters, and build a working model in Python from data collection through forecasting.

A hands-on guide to the silhouette score, covering the formula, interpretation ranges, a scikit-learn example, how to use it for choosing the right number of clusters, and how it compares to other clustering metrics.

Knowing why spurious relationships happen, from confounders to sampling bias, is what separates a real finding from a statistical coincidence.

Learn what the lognormal distribution is, how it relates to the normal distribution, and where it applies across finance, biology, and machine learning.

Compare stock market data APIs for backtesting, dashboards, screeners, and AI agents, with Python examples covering splits, reliability, and fundamentals.

Kernel density estimation is a nonparametric method for estimating the shape of a data distribution without assuming a fixed model. Learn the formula, bandwidth selection, and hands-on implementation in Python and R.

A practical walkthrough of the assumptions behind logistic regression, the diagnostics that catch violations in Python and R, and the alternatives to reach for when the assumptions don't hold.

A practical guide to spline regression, covering how piecewise polynomials and knots model nonlinear relationships, the main spline types, and how to fit them in Python and R.

A practical guide to GLMs - what they are, how their three components work together, and how to fit and interpret them in Python and R.

A detailed walkthrough of overfitting and underfitting in machine learning, including how to identify each failure mode, why it happens, and how to fix it through the bias-variance tradeoff.