Unified Software for Estimating Dynamic Production Functions and Machine Learning Models

Title	Unified Software for Estimating Dynamic Production Functions and Machine Learning Models
Authors	D. I. Khokha¹ ¹Yanka Kupala State University of Grodno
Annotation	Original software has been developed for constructing and comparing models based on dynamic production functions, regression machine learning models and recurrent neural networks. The system implements Cobb–Douglas, Wicksell and CES production functions with neutral scientific and technological progress and distributed lags of production factors; regression machine learning models, including linear regression, Lasso regression, Ridge regression, ElasticNet, random forest and gradient boosting; and recurrent neural networks based on RNN, LSTM and GRU layers. A computational experiment was performed using statistical data for BRICS countries and partner countries for 1990–2024.
Keywords	software system, production function, time lags, scientific and technological progress, machine learning, recurrent neural networks.
Citation	Khokha D. I. ''Unified Software for Estimating Dynamic Production Functions and Machine Learning Models'' [Electronic resource]. Mathematical modeling, numerical methods, and software systems: Collection of materials from the 12th All‑Russian Scientific Youth School‑Seminar named after E. V. Voskresensky (Saransk, July 21–24, 2026). - pp. 236-240. Available at: https://conf.svmo.ru/files/2026/papers/paper113.pdf. - Date of access: 25.09.2026.