Using Financial and Economic Leading Indicators to Predict Sales of Publicly Traded Companies
This article proposes a modeling procedure that combines time series and regression analysis for estimating sales
of publicly traded companies based on internal financial and economic leading indicators. First, this article proposes a data
transformation equation to improve linear relationships between preceding financial and economic variables and sales
performance. Second, based on these improved relationships, a modeling procedure that combines time series and regression
analysis is used to develop sales forecasting models for four sample construction companies. The out-of-sample forecasting
accuracy is evaluated using mean absolute percentage error (MAPE). The results show that the MAPE values in the
forecasting models range from 0.89% to 4.94% with an average of 2.68%, which outperforms a similar study that uses the
vector auto regression (VAR) model and the Litter man Bayesian vector auto regression (LBVAR) model.
Keywords - Sales Forecasting, Structural Model, Time Series Regression Model.