Temporal disaggregation

Temporal disaggregation, interpolation and extrapolation of time series. Methods: univariate (with or without indicators) and multivariate.

This library contains a complete set of MATLAB functions designed to perform temporal disaggregation of time series using a variety of techniques: methods without indicators (Boot-Feibes-Lisman, Stram-Wei, low-pass interpolation); methods with indicators using different approaches: optimization (Denton), static models (Chow-Lin, Fernandez and Litterman), dynamic models (Santos-Cardoso, Proietti), ARIMA models (Guerrero); and multivariate methods with indicators and transversal constraints (Denton, Rossi, Di Fonzo). The library contains also functions for balancing (proportional, RAS bi-proportional and Van der Ploeg) as well as an interface written in Visual Basic that allows its use in a spreadsheet environment. This library is intended for its use in production mode, easing the tasks of regular data compilation and short-term monitoring, and also for its use in research mode, allowing an in-depth exploration of the results and its internal mechanics.