Prediction of Gold Price Movement Considering the Number of Infected with the Covid 19

Jovana Stokanović Sević and Ana Jovancai Stakić

The European Journal of Applied Economics, 2022, vol. 19, issue 2, 71-83

Abstract: This paper aims to test several models and select the best one for predicting the price of gold on the world market for the next day, in five and ten days, taking into account the number of cases and deaths from the Covid-19 virus. These predictions can help decision-makers whether, at what point, and in what amount, it is best to invest in gold and gold-related financial instruments, relative to the projected price of gold from the model. The paper tests models called Decision tree, K-nearest neighbors, Linear regression model, and Support vector machines based on the information on gold prices and the number of cases and deaths from the Covid-19 virus. It will be seen in the paper that even models with only information on the price of gold give quite reliable predictions, but in unstable times like this, models that take into account the instability factor give more accurate predictions. The research aims to determine the optimal amount of information based on which the models will "learn" to give the most accurate possible result. This work’s data processing and models are done in Python.

Keywords: Gold price; COVID-19; Decision tree; K-nearest neighbors; Linear regression model; Support vector machine. (search for similar items in EconPapers)
Date: 2022
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journal.singidunum.ac.rs/files/2022-19-2/p ... ith-the-covid-19.pdf (application/pdf)
https://journal.singidunum.ac.rs/paper/prediction- ... ed-with-the-covid-19 (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:sng:ejae01:174

DOI: 10.5937/EJAE19-39258

Access Statistics for this article

More articles in The European Journal of Applied Economics from Singidunum University, Belgrade
Bibliographic data for series maintained by EJAE Editorial Office ().