Abstract:

This paper depends on research that was finished to become familiar with the connection between the cost of gold and a couple of significant variables, similar to the securities exchange, the cost of unrefined petroleum, the conversion scale of the rupee, expansion, and loan fees. Month to month cost measurements from January 2000 to December 2018 were utilized in the review. What's more, the information were isolated into two periods: period I, when the cost of gold rose from January 2000 to October 2011, and period II, when it tumbled from November 2011 to December 2018. These facts were broken down using four distinct ML calculations: linear regression, voting regression, gradient boosting regression, and random forest regression. There are significant areas of strength for a between the variables in period I and a powerless association in period II, it has been determined. During period I, these models fit the data well, but during period II, they don't. While gradient boosting regression is more accurate for each of the two time periods, random forest regression has a higher forecast accuracy over the entire time period.

Description:

Machine Learning, Regression, Prediction.