The Future of High-Cap Investments: The Challenges of Forecasting

Using census data, I undertook the task of predicting Target store locations by simulating a 2017 perspective and projecting forward to 2019. Despite relying on population growth data, my recommendations and actual store placements rarely aligned with high-growth areas, underscoring the complexity of accurate long-term forecasting.

Key Takeaways

• Short-Term Focus: Limit predictions to within two years for greater accuracy.

• Causal Models: Enhance forecasts with causal prediction models.

• Flexible Strategies: Help stakeholders create a step-stone plan that builds in opportunities to re-evaluate and pivot as circumstances change. In this example, using ship-from-store capabilities transforms brick-and-mortar stores into mini-warehouses, allowing them to serve a broader area beyond their immediate location. This significantly enhances the store’s value within the larger network.

• Clear Communication: Inform stakeholders about the inherent uncertainties of long-term forecasts and provide estimated ranges instead of point estimates.

Sample code here