MODEL LAB

🛡️

ACCT 311 · Predictive Analytics

You already described the wholesale customers. Now build models that explain relationships, make predictions, and interpret output responsibly. The Mandalorian guide reminds you: disciplined interpretation first, dramatic claims never.

🛡️ Mando says: trust the output, but verify the meaning.

Analyst name

Same UCI Wholesale Customers dataset · 440 customers · six annual spending categories · no time variable. This means the lab teaches cross-sectional prediction, not time-series forecasting.

🔓 Instructor mode unlocked · all missions and answer notes are available.

Official UCI dataset ↗Download full UCI dataset ↗

Important modeling context: each row is one customer and each spending field is annual spending in monetary units (m.u.). The dataset has no year-by-year history, so it supports cross-sectional modeling and segmentation—not time-series forecasting.

Channel

Customer channel: 1 = Horeca (hotel/restaurant/café), 2 = Retail.

Region

1 = Lisbon, 2 = Oporto, 3 = Other region.

Fresh

Annual spending on fresh products.

Milk

Annual spending on milk products.

Grocery

Annual spending on grocery products. This lab uses Grocery as the regression outcome (Y).

Frozen

Annual spending on frozen products.

Detergents_Paper

Annual spending on detergents and paper products.

Delicassen

Annual spending on delicatessen products.

Classroom preview · full dataset remains available from UCI

#	Channel	Region	Fresh	Milk	Grocery	Frozen	Detergents/Paper	Delicassen

MODEL LAB

Instructor padlock

Model Evidence Notebook

UCI Wholesale Customers · data reference