Analysis and forecasting of income statement account balances the dynamic interdependency and ARIMA approaches

This paper develops a dynamic simultaneous-equations model for analyzing and forecasting the account balances in the income statement of a firm. In the model, the income statement accounts play the role of the dependent variables that are jointly determined and explained by three types of exogenous variables: non-controllable, performance, and controllable. The model is estimated by the three-stage least-squares method using annual series of data for six firms during the period from 1936 or 1950 to 1981. The immediate, delayed, and cumulated impacts of an exogenous shock on the income accounts are analyzed and the implications for managerial decisions and strategies discussed. To run as a standard of comparison for the dynamic interdependency model, the Box-Jenkins approach is also used to develop an autoregressive integrated moving average (ARIMA) model for each of the accounts. Assessing the forecasting performance of the dynamic model against a naive model and the ARIMA and Elliott-Uphoff models for the 1982-4 period beyond the estimation period, we conclude that the dynamic model is a better representation of income statement accounts of the firm and increases forecasting accuracy.