Man vs. Machine in Predicting Successful Entrepreneurs: Evidence from a Business Plan Competition in Nigeria

Abstract

This paper compares the relative performance of man and machine in being able to predict outcomes for entrants in a business plan competition in Nigeria. The first human predictions are business plan scores from judges, and the second are simple ad hoc prediction models used by researchers. The paper compares these (out-of-sample) performances with those of three machine learning approaches. The results show that (i) business plan scores from judges are uncorrelated with business survival, employment, sales, or profits three years later; (ii) a few key characteristics of entrepreneurs such as gender, age, ability, and business sector do have some predictive power for future outcomes; (iii) modern machine learning methods do not offer noticeable improvements; (iv) the overall predictive power of all approaches is very low, highlighting the fundamental difficulty of picking winners; and (v) the models do twice as well as random selection in identifying firms in the top tail of performance.

Keywords: Private Sector Economics, Private Sector Development Law, Marketing, Labor Markets, Gender and Development, Educational Sciences, Educational Populations, Educational Policy and Planning - Textbook, Education for Development (superceded), Education For All