Affordable Access

Access to the full text

Business success prediction in Rwanda: a comparison of tree-based models and logistic regression classifiers

Authors
  • Kipkogei, Francis1
  • Kabano, Ignace H.1, 1
  • Murorunkwere, Belle Fille1, 2
  • Joseph, Nzabanita1, 1
  • 1 University of Rwanda, Kigali, Rwanda , Kigali (Rwanda)
  • 2 Rwanda Revenue Authority, Kigali, Rwanda , Kigali (Rwanda)
Type
Published Article
Journal
SN Business & Economics
Publisher
Springer International Publishing
Publication Date
Jul 16, 2021
Volume
1
Issue
8
Identifiers
DOI: 10.1007/s43546-021-00104-2
Source
Springer Nature
Keywords
Disciplines
  • Review
License
Yellow

Abstract

Businesses contribute immensely to economic growth. However, many enterprises started fail within a year of their operation. This study seeks to predict business success, elucidating on factors affecting the success of a business based on recent data for timely intervention. The study used Rwanda Revenue Authority data. Tree-based models were compared with logistic regression for the prediction of the business success. Log loss, Area under Receiver-Operating Characteristic Curve, accuracy, recall, and F1 score were used to evaluate the performance of each model in differentiating between successful and unsuccessful business. Tree-based ensemble models were more robust than other classifiers. However, gradient boosting was the most robust model. The results showed that the business industry (sector of the economy) is the most important factor determining business success. Other important factors are the nature of the business and type of ownership, duration of operation, and location of the business.

Report this publication

Statistics

Seen <100 times