Time Series Prediction with Direct and Recurrent Neural Networks
Journal Title: Turkish Journal of Forecasting - Year 2017, Vol 1, Issue 1
Abstract
This article presents a comparative study of the prediction of time series for the Consumer Price Index (CPI) using a recurrent neural network (RNN). For this, three models are designed for recurrent networks, with changes in backpropagation applied to allow them to incorporate the ARX (auto-regressive with external input) and NARX (nonlinear autoregressive with external input) models. We also present a third architecture, re-fed with the hidden layer, called ARXI, which is a special case of the Elman Network. Training is carried out for all networks and tests the ability to generalize them (identification stage), in order to select the best architectures of recurrent networks for prediction of the CPI. Following this stage, the models are validated by testing the extrapolation capacity of the networks, i.e., the presented data were not used during the training phase and obtain responses that indicate the capacity to predict the CPI at various times in the future (validation phase). We conclude that the NARX network shows the best performance and that the hybrid system proposed in [1] constitutes an excellent tool when minimal networks are required that make a series of perdition satisfactorily.
Authors and Affiliations
Lídio Mauro Lima De Campos
Keywords
Related Articles
- EP ID EP299789
- DOI -
- Views 128
- Downloads 0
A New Journal for Forecasting Research
The Turkish Journal of Forecasting (TJF) is an open access international journal and it is published semi-annually. The aim of the TJF is to procure a platform to integrate the research subjects and fields, and to bridge...
An Enhanced Neural-based Bi-Component Hybrid Model for Foreign Exchange Rate Forecasting
Foreign exchange rates are among the most important economic indices in the international monetary markets. For large multinational firms, which conduct substantial currency transfers in the course of business, being abl...
Time Series Prediction with Direct and Recurrent Neural Networks
This article presents a comparative study of the prediction of time series for the Consumer Price Index (CPI) using a recurrent neural network (RNN). For this, three models are designed for recurrent networks, with chang...
G-STAR Model for Forecasting Space-Time Variation of Temperature in Northern Ethiopia
Among many indicators of climate change, the temperature is a key indicator to take remedial action for world global warming. This finding provides application of space-time models for temperature data, which is selected...
Why and How Does Exponential Smoothing Fail? An In Depth Comparison of ATA-Simple and Simple Exponential Smoothing.
Even though exponential smoothing (ES) is publicized as one of the most successful forecasting methods in the time series literature and it is widely used in practice due to its simplicity, its accuracy can be affected b...