Using boosting for financial analysis and performance prediction: Application to s&p 500 companies, Latin American ADRs and banks

TY - JOUR

T1 - Using boosting for financial analysis and performance prediction

T2 - Application to s&p 500 companies, Latin American ADRs and banks

AU - Creamer, Germán

AU - Freund, Yoav

PY - 2010

Y1 - 2010

N2 - This paper demonstrates how the boosting approach can support the financial analysis functions in two ways: (1) As a predictive tool to forecast corporate performance, and rank accounting and corporate variables according to their impact on performance, and (2) As an interpretative tool to generate alternating decision trees that capture the non-linear relationship among accounting and corporate governance variables that determine performance. We compare our results using Adaboost with logistic regression, bagging, and random forests. We conduct 10-fold cross-validation experiments on one sample each of S&P 500 companies, American Depository Receipts (ADRs) of Latin American companies and Latin American banks. Adaboost results indicate that large companies perform better than small companies, especially when these companies have a limited long-term assets to sales ratio. Performance improves for large LAADR companies when the country of residence is characterized by a weak rule of law. In the case of S&P 500 companies, performance increases when the compensation for top officers is mostly variable.

AB - This paper demonstrates how the boosting approach can support the financial analysis functions in two ways: (1) As a predictive tool to forecast corporate performance, and rank accounting and corporate variables according to their impact on performance, and (2) As an interpretative tool to generate alternating decision trees that capture the non-linear relationship among accounting and corporate governance variables that determine performance. We compare our results using Adaboost with logistic regression, bagging, and random forests. We conduct 10-fold cross-validation experiments on one sample each of S&P 500 companies, American Depository Receipts (ADRs) of Latin American companies and Latin American banks. Adaboost results indicate that large companies perform better than small companies, especially when these companies have a limited long-term assets to sales ratio. Performance improves for large LAADR companies when the country of residence is characterized by a weak rule of law. In the case of S&P 500 companies, performance increases when the compensation for top officers is mostly variable.

KW - Adaboost

KW - Data mining

KW - Financial analysis

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=77955703769&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955703769&partnerID=8YFLogxK

U2 - 10.1007/s10614-010-9205-3

DO - 10.1007/s10614-010-9205-3

M3 - Article

SN - 0927-7099

VL - 36

SP - 133

EP - 151

JO - Computational Economics

JF - Computational Economics

IS - 2

ER -