Implementation multiple linear regresion in neural network predict gold price

Musli Yanto, Sigit Sanjaya, Yulasmi Yulasmi, Dodi Guswandi, Syafri Arlis

Abstract


The movement of gold prices in the previous period was crucial for investors. However, fluctuations in gold price movements always occur. The problem in this study is how to apply multiple linear regression (MRL) in predicting artificial neural networks (ANN) of gold prices. MRL is mathematical calculation technique used to measure the correlation between variables. The results of the MRL analysis ensure that the network pattern that is formed can provide precise and accurate prediction results. In addition, this study aims to develop a predictive pattern model that already exists. The results of the correlation test obtained by MRL provide a correlation of 62% so that the test results are said to have a significant effect on gold price movements. Then the prediction results generated using an ANN has a mean squared error (MSE) value of 0.004264%. The benefits obtained in this study provide an overview of the gold price prediction pattern model by conducting learning and approaches in testing the accuracy of the use of predictor variables.


Keywords


artificial neural networks; backpropagation; gold prices; multiple linear regression; prediction;

Full Text:

PDF


DOI: http://doi.org/10.11591/ijeecs.v22.i3.pp1635-1642

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics