Exploring Causal Decision Theory Approach with Quantile Regression

Author(s): Shenggang Li Originally published on Towards AI. Using AI and Causal Decision Theory to Prioritize Restocking: Balancing Demand, Inventory Risk, and Product Importance This member-only story is on us. Upgrade to access all of Medium. Photo by Louis Hansel on Unsplash When facing uncertainties, predictions guide us in selecting the best project from various options. Every decision aims for the best outcome, and Causal Decision Theory (CDT) provides a framework for choosing actions most likely to achieve that. What if we could apply CDT practically with the support of data and AI? This post studies a new approach that combines CDT with quantile regression, a machine-learning model that predicts demand at different levels. We use supply chain logistics as an example, where restocking choices are crucial. Should a business prioritize a product in high demand that’s nearly out of stock, or focus on items with stable demand? By forecasting product demands (or sales) — this approach helps weigh risks like stock outs and calculates a utility function to guide restocking decisions. Although demonstrated here with a supply chain example, this CDT-based model holds promise for other fields where data-driven, practical decisions are important. Let’s dive into a simple example to see how CDT helps us choose the best stock based on expected returns. Suppose we’re comparing three stocks: Microsoft (MSFT), Apple (AAPL), and Nvidia (NVDA). We’ll set up a table… Read the full blog for free on Medium. Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor. Published via Towards AI