Exploring datasets with missing variables in the context of multivariate analysis requires careful consideration to ensure the integrity and reliability of your results.

libscientific, compared with the most diffused ML packages (scikit-learn and so on), do not impute or exclude features with missing variables. Thanks to the NIPALS algorithm implemented in the most common and used multivariate algorithms (Principal Component Analysis, Partial Least-Squares, and Consensus Principal Component Analysis), skip the missing values, avoiding erroneous conclusions due to data alteration.

Is multivariate analysis a valid tool to consider in 2024, the AI era, LLM, GPT-4? The answer is Yes. Multivariate analysis is relevant and essential in various disciplines, including statistics, data science, economics, psychology, biology, chemistry, and general data science. Multivariate analysis helps organize data to uncover actionable insights for current problems and complex relations between data.