Trend time series modeling and forecasting with neural networks

Despite its great importance, there has been no general consensus on how to model the trends in time series data. Compared to traditional approaches, neural networks have shown some promise in time series forecasting. This paper investigates how to best model trend time series using neural networks. Four strategies (raw data, raw data with time index, detrending, and differencing) are used to model various simulated trend patterns (linear, nonlinear, deterministic, stochastic, and breaking trend). We find that with neural networks differencing often gives meritorious results regardless of the underlying DGPs. This finding is also confirmed by the real GNP series.