MSBAI LATAM Research Project

Understanding shops. Supporting shopkeepers.

LATAM Capstone Project Studies

Two connected studies exploring everyday micro-trade in Latin America—turning survey evidence into useful predictions, bringing shoppers into the picture, and examining how business change can be understood.

From real-world survey research to synthetic longitudinal experiments

Explore the four studies

Titles and full abstracts · Read in sequence or explore a topic

01Data preparation & predictive modeling

Latin American micro-retail · Survey research

Predicting Relative Commercial Activity in Latin American Micro-Retail

Abstract

This study develops a shopkeeper-based model that successfully predicts differences in commercial activity among Latin American micro-retail businesses. Using 1,291 interviews from Argentina, Brazil, Colombia, Mexico, and Peru, it turns a detailed account of shop operations, practices, and services into a common framework for comparison. Exploratory analysis describes the businesses and their priorities, while a carefully prepared activity measure combines customer traffic and spending relative to shops of comparable country, format, and size.

A 79-feature Elastic Net, selected through nested evaluation of four model families, achieved pooled development R² of 0.2992 and protected-test R² of 0.2540. On the 259 shops reserved for final evaluation, it reduced MAE by 10.18% and RMSE by 13.63% compared with the mean-target baseline, with MAE of 0.6915 and RMSE of 0.8530 on the log scale. Stable predictive patterns linked relative activity with staffing, digital connectivity, financial access, transaction services, product mix, and modernization. The study establishes a tested foundation for peer benchmarking, understandable business profiles, and more focused conversations between shopkeepers and those supporting them. It demonstrates how survey research can produce useful predictive tools for a diverse and underexplored area of regional commerce.

Read study
02Shopper augmentation & incremental value

Latin American micro-retail · Survey research

Do Shopper Surveys Add Predictive Value?

Abstract

This study evaluates whether linked shopper surveys improve prediction of relative shop performance beyond shopkeeper information alone. A conservative time-and-uniqueness rule linked 1,883 of 2,595 shopper interviews to 1,048 shops, after which shopper responses were aggregated to the shop level. A supplementary comparison on 841 linked development shops first tested four understandable shopper profiles: Deliberate Planners, Impulse Explorers, Loyal Optimizers, and Overspenders. Adding three nonredundant segment shares improved repeated-cross-validation RMSE from 0.7700 to 0.7676 and R² from 0.2634 to 0.2680. The richer 30-feature store-choice and desired-capabilities domain performed better, reaching RMSE 0.7610 and R² 0.2806; combining store choice with segments was numerically best at RMSE 0.7604 and R² 0.2818, although the additional gain was very small. In the original governed evaluation, the locked 109-feature augmented Elastic Net reduced protected-test MAE by 4.69%, reduced RMSE by 4.01%, and increased R² from 0.2154 to 0.2771 on 207 matched shops. Shopper segments therefore provide a compact way to describe customer mix, while detailed store-choice information remains the stronger and better-validated predictive addition.

Read study