Analysis of some cost forecasting techniques in management accounting: the case of a small or medium-sized enterprise

An essential point for SMEs and other big companies that strive for success concerns their expenses. By implementing new costing techniques, they mainly expect to improve their profitability, reduce their costs and improve their information system (Bright, Davies, Downes, & Sweeting, 1992). A company must identify the method to use to evaluate its costs according to some activities or cost nature but it is much more difficult to imagine how they will evolve in the future. It is not always possible for companies to make cost forecasts of good quality due to a lack of data or resources, especially for SMEs. Making good forecasts is however greatly valuable for decision making at the appropriate moment. Forecasts are therefore of great interest for those companies and numerous mathematical tools provided by statistics can help them in this task. The objective of this thesis is to study a concrete case of an SME wishing to leverage statistical methods to improve its cost forecasts. We aim at evaluating methods frequently used in business forecasting to figure out the potential utility for an SME to use them instead of its own methods. Hence, we will develop and evaluate several forecasting time series models as a comparison to the forecasts made by the company on a monthly and quarterly basis.