Predicting Multi Factor Productivity – A Comparative Analysis Between The Neural Network and Multiple Regression

Multi Factor Productivity (MFP) is widely accepted as a better indicator than Partial Productivity ratios like Labor Productivity or Material Productivity to gauge the performance and competitiveness of an industry. For calculation of MFP, the value of gross output or total value of production is needed which requires long time to consolidate. Hence predicting MFP without this detail, is very much essential to ensure proactive rather than reactive management of resources. Predicting MFP helps to identify the trends, to monitor, to control the consumption of resources and to suitably change the strategies. Using the conventional statistical technique Multiple Regression (MR) one model was built to predict MFP. Another model was built using Artificial Neural Network (ANN). The performance of the models were analyzed through Mean Percentage Error (MPE) and Mean Absolute Percentage Error (MAPE).