Behavioral time-series quant research with synthetic futures modeling and backtesting.

Time series analysis explores temporal patterns, trends, and dependencies in sequential data. Using R, users can model, forecast, and decompose series with packages such as forecast, tsibble, and fable, enabling robust analysis for economics, environmental data, and other applied domains.

A project leveraging LSTM and XGBoost to analyze and predict energy usage, identify inefficiencies, and optimize costs across buildings, providing actionable insights for energy management and savings.

Enhancing time series forecasting stability with a connected neural dynamical system approach