Demand forecasting is an age old concept being practiced by businesses and conventional models were designed to forecast based on historical data but businesses keep discovering newer internal and external influencers to demand which they might not have experienced in past to learn from. This raises a need for AI models to be designed to handle unprecedented exogenous factors to enable better forecasting to varied demand. We would like to share our experience of designing such models along with its impact at some of clients. We will be using google colab to share and execute the code snippets. Be ready with your google account to do hands-on with us.
a. Forecasting
- https://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.curve_fit.html
- https://docs.scipy.org/doc/scipy/reference/tutorial/optimize.html
- https://www.elegantjbi.com/blog/what-is-arimax-forecasting-and-how-is-it-used-for-enterprise-analysis.htm
- https://otexts.com/fpp2/
- CausalML: Python Package for Causal Machine Learning : https://arxiv.org/pdf/2002.11631.pdf
- https://blog.exploratory.io/an-introduction-to-causal-impact-analysis-a57bce54078e
- https://www.youtube.com/watch?v=GTgZfCltMm8
