Ratios determining of microenterprise profitability: an analysis based on multilinear regression models

Introduction: to determine the best multilinear regression model(s) of the Return on Assets and the Return on Equity of microenterprises in the commune of San Bernardo de Santiago de Chile; to strengthen their competitiveness in the industry, contributing to the strengthening of public policy in the sector, the source of information was the Mayor's Office, Directorate of Productive Development; Local Economic Development and College of Accountants Method: an analysis of financial ratios was developed, using predictor variables according to economic sector and size, consolidating 29 financial ratios derived from the balance sheets and income statements of the last 5 business years, discriminating collinearity by Pearson correlation and variance inflation factor (VIF> 10), then a normality analysis was developed through the Kolmogorov-Smirnoff and Shapiro - Wilk tests. generating 8 multiple linear regression models, evaluating the metrics R2, adjusted R2, F and P-Value Statistics, MAE (mean absolute error) RMSE (mean square error), to relate the measures of profitability. Results: 6 of the 8 models present high degrees of prediction, performance and level of adjustment, demonstrating statistical soundness of the ROA and ROE indices for the financial performance of microenterprises, segmented by sector and size, defining the impact of long-term debt, working capital, current amount, DuPont and acid index on profitability. Conclusions: The study constructed 6 statistically robust and highly predictive regression models to define the financial variables with the highest degree of impact on microenterprise profitability.