Time series analysis and forecasting
Hands-on implementation of classical time series models as part of advanced coursework.
Problem
Understand how classical time series models behave in practice, beyond theoretical definitions.
Approach
Implemented AR, MA, ARIMA, and ARIMAX models from scratch, including stepwise model selection and diagnostics for heteroskedasticity.
Validation
Compared model behavior across datasets and checked assumptions using residual analysis and standard statistical tests.
Outcome
A much more grounded understanding of time series modeling and its limitations in real data.