| Title: | Hybrid Time Series Forecasting Using Error Remodeling Approach |
| Version: | 0.1.0 |
| Author: | Tanujit Chakraborty [aut, cre, cph] |
| Description: | Method and tool for generating hybrid time series forecasts using an error remodeling approach. These forecasting approaches utilize a recursive technique for modeling the linearity of the series using a linear method (e.g., ARIMA, Theta, etc.) and then models (forecasts) the residuals of the linear forecaster using non-linear neural networks (e.g., ANN, ARNN, etc.). The hybrid architectures comprise three steps: firstly, the linear patterns of the series are forecasted which are followed by an error re-modeling step, and finally, the forecasts from both the steps are combined to produce the final output. This method additionally provides the confidence intervals as needed. Ten different models can be implemented using this package. This package generates different types of hybrid error correction models for time series forecasting based on the algorithms by Zhang. (2003), Chakraborty et al. (2019), Chakraborty et al. (2020), Bhattacharyya et al. (2021), Chakraborty et al. (2022), and Bhattacharyya et al. (2022) <doi:10.1016/S0925-2312(01)00702-0> <doi:10.1016/j.physa.2019.121266> <doi:10.1016/j.chaos.2020.109850> <doi:10.1109/IJCNN52387.2021.9533747> <doi:10.1007/978-3-030-72834-2_29> <doi:10.1007/s11071-021-07099-3>. |
| Maintainer: | Tanujit Chakraborty <tanujitisi@gmail.com> |
| License: | GPL-2 | GPL-3 [expanded from: GPL (≥ 2)] |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.2.3 |
| Imports: | forecast, nnfor, stats, WaveletArima, Metrics |
| Depends: | datasets |
| Suggests: | ggplot2 |
| NeedsCompilation: | no |
| Packaged: | 2023-04-08 12:58:46 UTC; mad-s |
| Repository: | CRAN |
| Date/Publication: | 2023-04-11 08:10:08 UTC |
Hybrid ARIMA ANN Forecasting Model
Description
Hybrid ARIMA ANN Forecasting Model
Usage
arima_ann(y, n, PI = FALSE)
Arguments
y |
A numeric vector or time series |
n |
An integer specifying the forecast horizon |
PI |
A logical flag (default = |
Value
The forecast of the time series of size n is generated along with the optional
output of confidence interval (PI = TRUE) for the forecast.
References
Zhang, G. P. (2003). Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing, 50, 159-175.
Chakraborty, T., Ghosh, I., Mahajan, T., & Arora, T. (2022). Nowcasting of COVID-19 confirmed cases: Foundations, trends, and challenges. Modeling, Control and Drug Development for COVID-19 Outbreak Prevention, 1023-1064.
Hybrid ARIMA ARNN Forecasting Model
Description
Hybrid ARIMA ARNN Forecasting Model
Usage
arima_arnn(y, n, PI = FALSE, ret_fit = FALSE)
Arguments
y |
A numeric vector or time series |
n |
An integer specifying the forecast horizon |
PI |
A logical flag (default = |
ret_fit |
A logical flag specifying that the fitted values of the model on the training set should be returned if true, otherwise, false (default) |
Value
The forecast of the time series of size n is generated along with the optional
output of fitted values (ret_fit = TRUE) and confidence interval (PI = TRUE)
for the forecast.
References
Chakraborty, T., Chattopadhyay, S., & Ghosh, I. (2019). Forecasting dengue epidemics using a hybrid methodology. Physica A: Statistical Mechanics and its Applications, 527, 121266.
Chakraborty, T., Ghosh, I., Mahajan, T., & Arora, T. (2022). Nowcasting of COVID-19 confirmed cases: Foundations, trends, and challenges. Modeling, Control and Drug Development for COVID-19 Outbreak Prevention, 1023-1064.
Examples
arima_arnn(y = datasets::lynx, n = 3)
Hybrid ARIMA WARIMA Forecasting Model
Description
Hybrid ARIMA WARIMA Forecasting Model
Usage
arima_warima(y, n, p = 5, q = 5, PI = FALSE, ret_fit = FALSE)
Arguments
y |
A numeric vector or time series |
n |
An integer specifying the forecast horizon |
p |
An integer indicating the maximum order of AR process. Default is 5. |
q |
An integer indicating the maximum order of MA process. Default is 5. |
PI |
A logical flag (default = |
ret_fit |
A logical flag specifying that the fitted values of the model on the training set should be returned if true, otherwise, false (default) |
Value
The forecast of the time series of size n is generated along with the optional
output of fitted values (ret_fit = TRUE) and confidence interval (PI = TRUE) for the forecast.
References
Chakraborty, T., & Ghosh, I. (2020). Real-time forecasts and risk assessment of novel coronavirus (COVID-19) cases: A data-driven analysis. Chaos, Solitons & Fractals, 135, 109850.
Chakraborty, T., Ghosh, I., Mahajan, T., & Arora, T. (2022). Nowcasting of COVID-19 confirmed cases: Foundations, trends, and challenges. Modeling, Control and Drug Development for COVID-19 Outbreak Prevention, 1023-1064.
Examples
arima_warima(y = datasets::lynx, n = 3)
Hybrid ETS ARNN Forecasting Model
Description
Hybrid ETS ARNN Forecasting Model
Usage
ets_arnn(y, n, PI = FALSE, ret_fit = FALSE)
Arguments
y |
A numeric vector or time series |
n |
An integer specifying the forecast horizon |
PI |
A logical flag (default = |
ret_fit |
A logical flag specifying that the fitted values of the model on the training set should be returned if true, otherwise, false (default) |
Value
The forecast of the time series of size n is generated along with the optional
output of fitted values (ret_fit = TRUE) and confidence interval (PI = TRUE)
for the forecast.
Examples
ets_arnn(y = datasets::lynx, n = 3)
Hybrid Random Walk ANN Forecasting Model
Description
Hybrid Random Walk ANN Forecasting Model
Usage
rw_ann(y, n, PI = FALSE)
Arguments
y |
A numeric vector or time series |
n |
An integer specifying the forecast horizon |
PI |
A logical flag (default = |
Value
The forecast of the time series of size n is generated along with the optional
output of confidence interval (PI = TRUE) for the forecast.
Hybrid Random Walk ARNN Forecasting Model
Description
Hybrid Random Walk ARNN Forecasting Model
Usage
rw_arnn(y, n, PI = FALSE, ret_fit = FALSE)
Arguments
y |
A numeric vector or time series |
n |
An integer specifying the forecast horizon |
PI |
A logical flag (default = |
ret_fit |
A logical flag specifying that the fitted values of the model on the training set should be returned if true, otherwise, false (default) |
Value
The forecast of the time series of size n is generated along with the optional
output of fitted values (ret_fit = TRUE) and confidence interval (PI = TRUE)
for the forecast.
Examples
rw_arnn(y = datasets::lynx, n = 3)
Summarized score of all the hybrid models implemented in this package
Description
Summarized score of all the hybrid models implemented in this package
Usage
summary_hybridts(train, test)
Arguments
train |
A numeric vector or time series object for training the hybrid models |
test |
A numeric vector or time series object for evaluating the hybrid models |
Value
A data frame where the rows represent the out-of-sample scores for each of the hybrid models and the columns represent the RMSE, MASE, and sMAPE scores.
Hybrid Theta ANN Forecasting Model
Description
Hybrid Theta ANN Forecasting Model
Usage
theta_ann(y, n, PI = FALSE)
Arguments
y |
A numeric vector or time series |
n |
An integer specifying the forecast horizon |
PI |
A logical flag (default = |
Value
The forecast of the time series of size n is generated along with the optional
output of confidence interval (PI = TRUE) for the forecast.
Hybrid Theta ARNN Forecasting Model
Description
Hybrid Theta ARNN Forecasting Model
Usage
theta_arnn(y, n, PI = FALSE, ret_fit = FALSE)
Arguments
y |
A numeric vector or time series |
n |
An integer specifying the forecast horizon |
PI |
A logical flag (default = |
ret_fit |
A logical flag specifying that the fitted values of the model on the training set should be returned if true, otherwise, false (default) |
Value
The forecast of the time series of size n is generated along with the optional
output of fitted values (ret_fit = TRUE) and confidence interval (PI = TRUE) for the forecast.
References
Bhattacharyya, A., Chakraborty, T., & Rai, S. N. (2022). Stochastic forecasting of COVID-19 daily new cases across countries with a novel hybrid time series model. Nonlinear Dynamics, 1-16.
Bhattacharyya, A., Chattopadhyay, S., Pattnaik, M., & Chakraborty, T. (2021, July). Theta Autoregressive Neural Network: A Hybrid Time Series Model for Pandemic Forecasting. In 2021 International Joint Conference on Neural Networks (IJCNN) (pp. 1-8). IEEE.
Examples
theta_arnn(y = datasets::lynx, n = 3)
Hybrid WARIMA ANN Forecasting Model
Description
Hybrid WARIMA ANN Forecasting Model
Usage
warima_ann(y, n, p = 5, q = 5, PI = FALSE)
Arguments
y |
A numeric vector or time series |
n |
An integer specifying the forecast horizon |
p |
An integer indicating the maximum order of AR process. Default is 5. |
q |
An integer indicating the maximum order of MA process. Default is 5. |
PI |
A logical flag (default = |
Value
The forecast of the time series of size n is generated along with the optional
output of confidence interval (PI = TRUE) for the forecast.
References
Chakraborty, T., Ghosh, I., Mahajan, T., & Arora, T. (2022). Nowcasting of COVID-19 confirmed cases: Foundations, trends, and challenges. Modeling, Control and Drug Development for COVID-19 Outbreak Prevention, 1023-1064.
Hybrid WARIMA ARNN Forecasting Model
Description
Hybrid WARIMA ARNN Forecasting Model
Usage
warima_arnn(y, n, p = 5, q = 5, PI = FALSE, ret_fit = FALSE)
Arguments
y |
A numeric vector or time series |
n |
An integer specifying the forecast horizon |
p |
An integer indicating the maximum order of AR process. Default is 5. |
q |
An integer indicating the maximum order of MA process. Default is 5. |
PI |
A logical flag (default = |
ret_fit |
A logical flag specifying that the fitted values of the model on the training set should be returned if true, otherwise, false (default) |
Value
The forecast of the time series of size n is generated along with the optional
output of fitted values (ret_fit = TRUE) and confidence interval (PI = TRUE) for the forecast.
References
Chakraborty, T., Ghosh, I., Mahajan, T., & Arora, T. (2022). Nowcasting of COVID-19 confirmed cases: Foundations, trends, and challenges. Modeling, Control and Drug Development for COVID-19 Outbreak Prevention, 1023-1064.
Examples
warima_arnn(y = datasets::lynx, n = 3)