A Nonlinear Extension of the Nber Model for Short-Run Forecasting of Business Cycles

Abstract

To avoid the pitfalls of the widely used NBER model, in this paper we have adopted neural networks to forecast business cycles. We find that our model has overcome some of the main deficiencies of the classical leading indicators model: first, the model was able to correctly forecast all reference points in in-sample and out-of-sample data; second, the model can forecast the future value of reference series; and third, the model has a constant forecast horizon. Sensitivity analysis suggests there are some nonlinear relationships between the reference variable and selected leading indicators. This explains why we were able to improve the forecasting performance of the original model. Copyright 2005 Economic Society of South Africa.

Predicting Efficiency in Angolan Banks: A Two‐Stage TOPSIS and Neural Networks Approach

“…Artificial neural networks are one of the frequently used learning models for prediction (Jagric and Strasek, ; Ledolter, ). An artificial neural network is inspired by the structure of biological neural networks where neurons are interconnected and learned from experience.…”

The Properties of Cycles in South African Financial Variables and Their Relation to the Business Cycle

“…Consequently, the traditional approach evades the use of However, even in this context, the evaluation of leading indicators continues to be based on goodness-of-fit measures (see Moolman (2003) for the use of the "pseudo 12 Alternative non-linear specifications, such as the neural network, have been proposed to deal with the forecasting failures of the more conventional non-linear models. In a recent contribution, Jagric and Strasek (2005) claim that their neural network model of the Slovenian economy is able to forecast Slovenian industrial production accurately -even in the presence of business cycle turning points.…”