Macro-accounting Explanation with Emphasis on the Importance of Accounting Data in Inflation Modeling

Abstract

Objective: The aim of this study was to explain the importance of accounting data information in forecasting inflation rates, using a sample of 90 large companies listed in the Tehran stock exchange during 1385-1395 (1980 year-company). Method: Given the complex and nonlinear properties of inflation in this study, we relied on the predictive power of various artificial intelligence models including neural networks, genetic algorithms and particle swarm optimization. To have contribution to macro-accounting knowledge, a number of accounting variables were selected and their explanatory power was tested in forecasting two inflation rates of producer price index and consumer price index. Results: The findings indicated that the hybrid model of artificial neural networks, genetic algorithms and particle swarm optimization (HANGAPSO) are more accurate in predicting the inflation rates than other models. The model results, using accounting variables, also showed that the forecast error of producer price index is lower than the forecast error of consumer price index. Conclusion: In sum, the results of this study verify the importance of accounting information at macroeconomic level, and that this information should be used in macro-level decision-making.

Keywords

• accounting information
• economic prediction
• artificial intelligence models