Title: | Time Series Forecasting using ARIMA-ANN Hybrid Model |
---|---|
Description: | Testing, Implementation, and Forecasting of the ARIMA-ANN hybrid model. The ARIMA-ANN hybrid model combines the distinct strengths of the Auto-Regressive Integrated Moving Average (ARIMA) model and the Artificial Neural Network (ANN) model for time series forecasting.For method details see Zhang, GP (2003) <doi:10.1016/S0925-2312(01)00702-0>. |
Authors: | Ramasubramanian V. [aut, ctb], Mrinmoy Ray [aut, cre] |
Maintainer: | Mrinmoy Ray <[email protected]> |
License: | GPL-3 |
Version: | 0.1.0 |
Built: | 2024-12-18 06:29:39 UTC |
Source: | CRAN |
The ARIMAANN function fit ARIMA-ANN hybrid model for time series data.
ARIMAANN(data,h)
ARIMAANN(data,h)
data |
Input univariate time series (ts) data. |
h |
The forecast horizon. |
This package allows you to fit the ARIMA-ANN hybrid model.
Test_Result |
Checking the suitability of data for hybrid modelling |
ARIMA coefficients |
Coefficients of the fitted ARIMA |
pvalues |
pvalues of the fitted ARIMA model |
ANN Summary |
Summary of the fitted ANN model on residuals obtained from the fitted ARIMA model |
MAPE |
Mean Absolute Percentage Error (MAPE) of the fitted hybrid model |
MSE |
Mean Square Error (MSE) of fitted hybrid model |
fitted |
Fitted values of hybrid model |
forecasted.values |
h step ahead forecasted values employing hybrid model |
Ramasubramanian V., Mrinmoy Ray
Zhang, G. P.Time series forecasting using a hybrid ARIMA and neural network model Neurocomputing, 50 (2003), pp. 159-175.
auto.arima, nnetar
data=lynx ARIMAANN(data,5)
data=lynx ARIMAANN(data,5)