Forecasting inflation rate in Malaysia using Artificial Neural Network (ANN) Approach / Muhammad Athir Mohd Junaidi and Siti Aishah Mohd Shafie
Forecasting is very important for planning and decision-making in all fields to predict the conditions and cases surrounding the problem under study before making any decision. Hence, many forecasting methods have been developed to produce accurate predicted values. Inflation rates provide appropria...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Book Section |
| Language: | English |
| Published: |
Faculty of Computer and Mathematical Sciences
2023
|
| Subjects: | |
| Online Access: | https://ir.uitm.edu.my/id/eprint/93959/1/93959.pdf https://ir.uitm.edu.my/id/eprint/93959/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Forecasting is very important for planning and decision-making in all fields to predict the conditions and cases surrounding the problem under study before making any decision. Hence, many forecasting methods have been developed to produce accurate predicted values. Inflation rates provide appropriate and timely information about the trend changes, which affect the economy of Malaysia because of the different uses in many ways. It can be used as an economic indicator that is beneficial for policy makers, investors, and consumers to make a planning and decision. It also used as a supplement for statistical chains to predict future values rate to make sure that the data accurately reflect the pattern of the inflation rate in Malaysia. Therefore, the main objective of this study is to construct an inflation rate model for Malaysia and make a prediction of inflation rate for upcoming six months in 2023 by using artificial neural network (ANN). The proposed ANN model consists of an input layer, hidden layer, and output layer, while it applies the tangent function (TanH) as a testing and validation algorithm in the hidden layer. Finally, the predicted values of inflation rate are compared with the measured values. The proposed ANN model with four hidden nodes is more efficient than other models in predicting inflation rate after considering parsimonious architecture. The obtained Coefficient of Determination and Root Mean Squared Error (RMSE) values using the ANN method is 0.6399 and 0.9879 respectively. This study presents a new model for forecasting inflation rate values that are beneficial for many parties. The model was used to predict upcoming months then compared to the actual data. |
|---|