ML Paper

Abstract:
Historically, among the various other forms of payment, gold has been used for trade transactions
around the globe. Gold has been kept as a national reserve and it reflects the financial strength of a
country. It has been held with central banks of many countries to guarantee the re-payment of foreign
debts and also to control inflation. Besides the government, huge number of companies as well as
individuals have also invested in gold. In Asian countries, gold ornaments are presented as gifts in
marriages, etc. However, there is always an uncertainity in the changes of gold price. In this paper, we
aim to predict the price of gold using the various algorithms in machine learning. The performance of
world’s leading economies have an impact on the gold rates, so we can use the market variables in the
algorithm for the training of the data which will improve the accuracy of prediction. The prediction
model can be beneficial for central banks and investors. The prediction model considers parameters of
Oil price,EUR/USD,SP200 values.
Keywords:Gold,USD,EUR,Random Forest,Correaltion
INTRODUCTION:
One of the valuable metals is gold. It's been used as a form of payment, as well as for jewellery and
other reasons. Because of its limited availability and great value, it is employed as a medium of
exchange. Due to its scarcity and complexity of mining, this metal became a prized commodity. It also
shows the country's economic strength, which prompted many businesses and people to begin investing
in gold reserves. Many people saw gold as an interesting investment due to its rising value.
The price of gold is influenced by a variety of factors, resulting in price volatility. Inflation, demand and
supply, and political issues are just a few of the elements at play. When inflation rises, it clearly drives
the gold price higher, whereas when there is a scarcity of an item, its price rises. Furthermore, when
countries are concerned that the value of the dollar, which is the world's main currency, would collapse,
the gold price will rise as a result of the increased demand for gold. Other literature has referred to it as
a safe haven during financial crises because of its significance.
The price of gold fluctuates wildly and is difficult to predict. The graph depicts the fluctuation of the gold
price over a 41-year period, from 1979 to December 2019. Nonetheless, the gold price may be predicted
in advance, allowing for future decision-making.
The rising value of gold, combined with the volatility and price declines in other sectors such as capital
markets and real estate markets, has attracted an increasing number of investors to gold as a viable
investment. However, the price of gold has recently experienced tremendous volatility, making gold
investments riskier. There is concern about whether these high prices can be sustained and when they
will revert. Despite the fact that a number of research have been conducted to examine the relationship
between the price of gold and various economic indicators. In order to determine the dynamic
consequences of these relationships, a study to disclose the influence and impact of various macro-
economic factors on the price of gold in the current circumstances is still regarded useful. As a result,
the purpose of this article is to investigate the relationship between gold price and a number of
economic and market variables. Understanding such a relationship will aid not just monetary
policymakers, but also investors, fund managers, and portfolio managers in making better market
investment decisions.
Several studies have been conducted to examine the relationship between the gold price and other
economic indicators. Investors can make better selections by understanding the relationship between
these elements. We employ Random Forest Machine learning algorithm which is widely used in
regression and classification problems. Random Forest Algorithm can handle the data set containing
continuous variables as in the case of regression and categorical variables in case of classification.We
also correlate the price of gold with SP200,USO(oil price),SLV(silver price),EUR/USD.
Literature Survey
V.K.F.B.Rebecca Davis in the paper “Modeling and Forecasting of Gold Prices on Financia Markets” using
Autoregressive Moving Average (ARMA) model, which is a mathematical model commonly used to
analyze time series data. Estimated monthly gold prices for more than a decade. The model gained the
accuracy of 66.67%.
Iftikhar ul Sami and Khurum NazirJunejo in their paper entitled “Predicting Future Gold Scales Using
Machine Learning Approach ”uses Artificial Neural Networks (ANN) to predict gold prices. Data for this
study are collected from different sources over a period of eleven years. This data included variables
such as oil prices, S&P 500 index, USD exchange rates and other economic fluctuations.
D Makala and Z Li in their paper “Prediction of gold price with ARIMA and SVM” using data collected
from World Gold Council containing daily gold prices from January 1979 to December 2019. This study
uses Autoregressive Integrated Moving Average (ARIMA) method and SVM gold price forecast. Accuracy
achieved with the ARIMA model is less than the accuracy obtained by SVM.
"Big Data Analytics Predicting Gold Price Based on Decision Tree Algorithm and Vector Reduction
Support" by Navin And Dr.G.Vadivu using decision tree and vector regression support algorithm to
predict and predict gold prices using data containing five attributes namely open, close, low, high and
gold values. In this study, the decision tree is best suited for feature selection while SVR is best suited for
large training data.
Sima P.Patil, V.M.Vasava and G.M.Poddar in their paper entitled “Gold Market Analyzer uses Selection
based algorithm has given us insight into the factors that affect the price of gold. These features include
the future of silver, copper futures, oil futures, EUR-USD, S&P / TSX combined. They use algorithms like
SVM, Logistic regression for forecast. Logistic regression model gained 61.92% accuracy over SVM
model.
R. Hafezi *, A. N. Akhavan, "Forecasting Gold Price Changes: Use of Artificial Performance Sensor
Network ", found that the exchange rate is major problem in money-related markets.
Therefore, build intuitive strong selection model is fundamental to finance experts. As gold has shown a
unique volume to accelerate the diversity of inflammation, the leading agents use gold as a cost control
method. This paper is trying to propose a clever model developed by artificial sensory systems (ANNs) to
expand future cost of gold. The proposed smart system is provided with a meta-heuristic calculation
called BAT calculation to makes ANN fit for the following variants. Systematic model compared to
plausible distributed paper too other important models, for example, Autoregressive
Integrated Moving Average (ARIMA), ANN, Adaptive Neuro Fuzzy Inference System (ANFIS), Multilayer
Perceptron (MLP) Neural Network, Radial Basis Function (RBF) Neural Network and Neural Regression
Networks Generalized (GRNN).

ML Paper

Uploaded by

Copyright:

Available Formats

ML Paper

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ML Paper

Uploaded by

Copyright:

Available Formats

Abstract:

You might also like