Forecasting in Financial Data Context
Asian Journal of Economic Modelling
This paper aims to compare the application of most common forecasting techniques within financial data context such as regression analysis and artificial neural network, according to the most popular forecasting efficiency indices. Findings depicted that robust regression has an advantages over the least square regression and ANN, in financial data analysis because of outliers derived from business cycles. To this aim, relationship between earning per share, book value of equity per share and
... ity per share and share price as price model and earnings per share, annual change of earning per share and return of stock as return model scrutinized implementing robust and least square regressions as well as ANN. Based on results, it can be concluded that the robust regression can provide better and more reliable analysis owing to eliminating or reducing the contribution of outliers and influential data. Therefore, robust regression outperforms OLS and ANN, and can be recommended reaching more precise analysis in financial data context. This study contributes in the existing literature of financial forecasting models. Since it compare two common forecasting techniques with robust regression analysis based on the most important forecasting indices and clarified that robust analysis outperforms the other techniques because of the existence of outliers in financial data context.