See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/362491642

Gold Price Prediction using Machine Learning

Article in International Journal of Environmental Engineering · August 2022

DOI: 10.55041/IJSREM15027

CITATIONS READS

11 15,446

5 authors, including:

Jyoti Kharade
Institute of Management and Information Technology
54 PUBLICATIONS 38 CITATIONS

SEE PROFILE

All content following this page was uploaded by Jyoti Kharade on 05 August 2022.

The user has requested enhancement of the downloaded file.

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 06 Issue: 06 | June - 2022 Impact Factor: 7.185 ISSN: 2582-3930

Gold Price Prediction using Machine Learning

Rushikesh Ghule, Abhijeet Gadhave
Under the Guidance of Prof.Manish Dubey, Dr.Jyoti Kharade
Bharati Vidyapeeth's Institute of Management & Information Technology
Navi Mumbai

1. Abstract:- investors have conjointly been interested in this

valuable and invested vast amounts in it.
We predict future gold rates supported twenty Recently, rising world economies, like China,
two market variables victimization machine Russia, and India are massive buyers of gold,
learning technique.One machine learning whereas USA, South Africa, and Australia area
algorithm, random forest regression were used in unit among the large vender of this goods .
analyzing these knowledge[1]. Historically, gold Chinese and Indian traditional events conjointly
was used for supporting trade transactions around have an effect on the worth of the gold. in this time
the world besides alternative modes of payment. more money is poured for purchase of this
Various states maintained and increased their gold goods[4]. Small investors conjointly realize this
reserves and were recognized as rich and goods for safe investment rather than alternate
progressive states. In present times, precious investment choices, that bear in-built investment
metals like gold area unit control with central risks. Internal monetary conditions of the
banks of all countries to make sure re-payment of aforementioned countries play an important role
foreign debts, and conjointly to control inflation. for setting spot rates for gold. Gold is another plus
Moreover, it conjointly reflects the Imoney that is being thought-about as a lovely investment
strength of the country[2]. Besides government avenue by several investors thanks to its
agencies, varied transnational firms and people increasing worth and therefore the space of
have conjointly invested with in gold reserves. In usage[15]. Investors preference for gold as a
ancient events of Asian countries, gold is in protecting plus will increase thanks to their
addition presented as gifts/souvenirs and in negative expectations regarding things within the
marriages, gold ornaments are conferred as gift in developed interchange markets and therefore the
Republic of India. capital markets. Gold is additionally thought-
KEYWORDS: Price prediction, Machine about to be “the plus of ultimate instance” i.e. is
Learning, Supervised Learning, Linear that the plus investors place confidence in, once
Regression, Python , Power Bi , Tableau. the developed world capital markets aren't
capable to supply fascinating profitability. so it is
same that investors see gold as a tool to hedge
2. Introduction:- against the fluctuations in alternative markets[5].
Gold could be a valuable, thus like all alternative
Investing in gold has developed over a amount merchandise, gold’s value ought to rely on offer
of your time in conventional forms by buying and demand.
jewellery or through modern strategies, either by
buying gold coins and bars (which area unit 3. Literature Review:-
already accessible in scheduled
banks).Historically, gold had been used as a sort Gold is that the alone and that retains its worth
of currency in various components of the planet as even through the political and economic[14].
well as USA[3] . In recent times also, gold has downturn. The gold values unit of
maintained its worth and has been used as a means measurement usually directly connected to
for assessing the monetary strength of a rustic. Big different resources. Future gold worth forecast is

© 2022, IJSREM | www.ijsrem.com | Page 1

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 06 Issue: 06 | June - 2022 Impact Factor: 7.185 ISSN: 2582-3930

that the investors' alert mechanism thanks to benefit, especially to housing developers and
unpredictable market risk. Therefore,precise researchers, to ascertain the most significant
prognostication of gold prices is needed to predict attributes to determine house prices and to
the market patterns[6]. several machine intelligent acknowledge the best machine learning model to
techniques for gold prognostication applications be used to conduct a study in this field.
square measure noted over the past decade. the
link between gold value and costs of different Used Car Price Prediction:
commodities particularly petroleum has This research based on prediction price of used
conjointly been extensively studied. however the cars by using Machine Learning Algorithms such
results from these studies square measure found to as Lasso Regression, Multiple Regression and
be contradicting. a number of the studies on the Regression trees, we will try to develop a
factors influencing gold value and numerous statistical model which will be able to predict the
techniques used for learning these price of a used car, based on previous consumer
relationships[7]. data and a given set of features. We will also be
Some of the research papers on Price comparing the prediction accuracy of these
prediction are listed below: models to determine the optimal one.
Stock Closing Price Pridiction :
4. Dataset:-
This research based on accurate prediction of
Data for this study is collected from (kaggle.com)
stock market returns is a very challenging task due
Dec 2011 to September 2016 from numerous
to volatile and non-linear nature of the financial
sources. information for attributes, such as Oil
stock markets. With the introduction of artificial
worth, NYSE, normal and Poor’s (S&P) five
intelligence and increased computational
hundred index, US Bond rates (10 years),
capabilities, programmed methods of prediction
EuroUSD exchange rates were gathered[8]. Data
have proved to be more efficient in predicting
of the many government central banks and 5 giant
stock prices. In this work, Artificial Neural
companies that have invested with
Network and Random Forest techniques have
Brobdingnagian amounts in gold have conjointly
been utilized for predicting the next day closing
been collected. worth of precious metals
price for five companies belonging to different
throughout this era is also enclosed within the
sectors of operation
analysis. Table I lists the net sources from that this
Bitcoin Price Prediction : information was extracted[9]. Table II lists of
In this paper, we use the LSTM version of these attributes. The price of gold that we tend to
Recurrent Neural Networks, pricing for Bitcoin. try to predict is taken in US Dollar. plenty of
To develop a better understanding of its price cleansing and preprocessing was performed on the
influence and a common view of this good dataset. the matter of missing values was handled
invention, we first give a brief overview of in appropriate manner to finish the dataset.Gold
Bitcoin again economics costs amendment on everyday and are full of
major world events. Current gold rates ar a lot of
House Price Prediction: beyond a few years past.
This research is carried out to analyze the relevant
attributes and the most efficient models to
forecast the house prices. The findings of this
analysis verified the use of the Artificial Neural
Network, Support Vector Regression and
XGBoost as the most efficient models compared
to others. Moreover, our findings also suggest
that locational attributes and structural attributes
are prominent factors in predicting house
prices. This study will be of tremendous

© 2022, IJSREM | www.ijsrem.com | Page 2

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 06 Issue: 06 | June - 2022 Impact Factor: 7.185 ISSN: 2582-3930

Figure-2.Methodology used for price prediction

Figure-1.Data Sample using machine learning

5. Machine Learning Models:- The specialists then collect, pick, plan, pre-
process and convert this information. Upon
completion of this stage, the specialists begin
We use 2 milliliter models, particularly building prognosticative models[12]. A model
Random Forest, and linear regression. Random that predicts costs at the utmost preciseness rate
forest is a Supervised Machine Learning would be chosen to drive a tool or program. and
Algorithm that is used widely in Classification therefore the worth prediction perform system
and Regression problems[13]. In addition to the would possibly appear as if this:
input and output layers, they carries with it one or
1)Statement on drawback.
more hidden layers of neurons that try and learn
non-linear decision boundaries that separate 2)Knowing the peculiarities of economies.
totally different categories of information[10]. It respondent the question: What variables square
can also be accustomed predict continuous valued measure poignant artefact / product / service
attributes like gold costs in our case. Fig. four prices?
depicts a sample ANN. Linear regression (LR) is 3)Gathering, preparing, and preprocessing
AN approach utilized in statistics to model information.
relationship between dependent (class variable)
and one or additional freelance variables 4)Testing and modelling.
(attributes). regression can be used for predicting 5)Deploying a model into a code or
continuous valued attribute. We use the application
implementation of LR and ANN that's provided
by RapidMiner tool[11]. each models area unit
optimized exploitation the RMSE performance Machine learning algorithms square measure
live. oftentimes classified as supervised or
unsupervised.

© 2022, IJSREM | www.ijsrem.com | Page 3

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 06 Issue: 06 | June - 2022 Impact Factor: 7.185 ISSN: 2582-3930

7. Steps:-
1) Importing libraries
2) Data Understanding
3) Data Pre-processing & EDA
4) Checking Missing Values
5) Data Visualization
6) Statistical Measures (Mean, Standard
deviation, Kurtosis
7) Model Building
8) Project Deployment
Figure-3. Machine Learning Methods

7.1 Imported Libraries:-

6. Problem Formulation:-

From the above comprehensive literature

review on different prediction methods in
business applications utilizing analytical
intelligent techniques throughout the past
decades, it has been noted that the following
problems are found when carrying out prediction
processes for the business applications considered
– foreign exchange rate prediction, stock market
price prediction, gold price prediction:
1)Unattained Scalability
2)Premature Network Convergence model
3)Trapping ourselves in local and global
optima
4)Stilling Figure-4. Main Imported Libraries
5)Large prognostic bias
6)Overlapping computing energy 7.2 Data Understanding:-
7)The computational load of the algorithms
increased
8)No assurance on system's interpretability

Figure-5. Data Understanding

© 2022, IJSREM | www.ijsrem.com | Page 4

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 06 Issue: 06 | June - 2022 Impact Factor: 7.185 ISSN: 2582-3930

7.3 Data Pre-processing And EDA:-

Figure-5. Gold Price By Months

Figure-6 . Column Deletion

Figure-6. Gold Prices Lag Plot

7.5 Statistical Measures (Mean,

Standard-deviation, Kurtosis

Figure-7. Upsampling And Interpolation Coding :

7.4 Data Visualization:-

Data visualization is the graphical

representation of information and data. By
using visual elements like charts, graphs, and
maps, data visualization tools provide an
accessible way to see and understand trends,
outliers, and patterns in data.

© 2022, IJSREM | www.ijsrem.com | Page 5

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 06 Issue: 06 | June - 2022 Impact Factor: 7.185 ISSN: 2582-3930

Output: 8. Final Results:-

Figure-9. Gold Trend Prediction

Figure-7. Mean , Standard deviation and Kurtosis of

Gold Prices

7.6 Model Building:-

The model building process involves setting
up ways of collecting data, understanding and
paying attention to what is important in the Figure-10 . Dataset Information
data to answer the questions you are asking,
finding a statistical, mathematical or a
simulation model to gain understanding and
make predictions.

Figure-11. Gold Price Prediction Data

Visualization

Figure-8. All Models And Their RMSE Value

We use Random Forest for model building

because it’s gives high accurancy and As we
can see, the model's R-square is 97 % .

Figure-12 . Monthly Prediction

© 2022, IJSREM | www.ijsrem.com | Page 6

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 06 Issue: 06 | June - 2022 Impact Factor: 7.185 ISSN: 2582-3930

score close to 100 per cent indicates that the Gold

ETF prices are well explained by the given model.
3)Results show that proposed Random forest
method-machine learning beats customary and
current predicting models

10. References:-
[1] Xiaohui Yang, "The Prediction of Gold
Price Using ARIMA Model", 2nd International
Conference on Social Science, Public Health
and Education 2019.
Figure-12 . Accurate Predictions With 99.69 %
Accuracy
[2] Manjula K. A., Karthikeyan P, "Gold Price
Prediction using Ensemble based Machine
Learning Techniques", Third International
Conference on Trends in Electronics and
Informatics, 2019.

[3] Mrs. B. Kishori 1, V. Preethi, "Gold Price

forecasting using ARIMA Model", International
Journal of Research, 2018.

[4] R. Hafezi* , A. N. Akhavan, "Forecasting

Figure-13 . Daily Gold Prices Index
Gold Price Changes: Application of an Equipped
Artificial Neural Network", AUT Journal of
Modeling and Simulation, 2018.
9. Conclusion:-
This study by using machine learning [5] Xiaohui Yang, "The Prediction of Gold Price
algorithms to accurately predict the gold prices Using ARIMA Model", 2nd International
and when to sell them and purchase them. This Conference on Social Science, Public Health and
research was done in order to clarify the gold Education (SSPHE 2018).
ETF price predictions using machine learning
using Python. The research was carried out for
data between April 2020 and Jan 2022. The [6] Shian-Chang Huang and Cheng-Feng Wu,
results on proposed model is as per the Energy Commodity Price Forecasting with Deep
following: Multiple Kernel Learning, MDPI Journal, 2018.

1)It is concluded that machine learning algorithms

with Random Forest analysis are very useful in [7] Wedad Ahmed Al-Dhuraibi and Jauhar Ali,
gold price prediction. "Using Classification Techniques to Predict Gold
2)It is concluded that, the model's R-square is 97 Price Movement", 4th International Conference
percent. R-squared is usually 0 to 100 per cent. A on Computer and Technology Applications, 2018.

© 2022, IJSREM | www.ijsrem.com | Page 7

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 06 Issue: 06 | June - 2022 Impact Factor: 7.185 ISSN: 2582-3930

[8] Iftikharul Sami and KhurumNazirJunejo, Software Engineering and Computer Systems
"Predicting Future Gold Rates using Machine (ICSECS), 2015.
Learning Approach", International Journal of
Advanced Computer Science and Applications,
2017.

[9] NalinipravaTripathy, "Forecasting Gold Price

with Auto Regressive Integrated Moving Average
Model", International Journal of Economics and
Financial Issues, 2017.

[10] K. R SekarManav Srinivasan, K. S.

Ravichandran and J. Sethuraman, "Gold Price
Estimation Using A Multi Variable Model",
International Conference on Networks &
Advances in Computational Technologies, 2017.

[11] Sima P. Patil, Prof. V. M. Vasava, Prof. G.

M. Poddar, " Gold Market Analyzer using
Selection based Algorithm", International Journal
of Advanced Engineering Research and Science,
2016.

[12] S. Kumar Chandar, M. Sumathi and S. N.

Sivanadam, "Forecasting Gold Prices Based on
Extreme Learning Machine", International
Journal of Computers Communications &
Control, 2016. International Journal of Trend in
Scientific Research and Development (IJTSRD)
@ www.ijtsrd.com eISSN: 2456-6470 @ IJTSRD
| Unique Paper ID – IJTSRD33143 | Volume – 4 |
Issue – 5 | July-August 2020 Page 1456

[13] NurulAsyikin Zainal and ZurianiMustaffa,

"Developing A Gold Price Predictive Analysis
Using Grey Wolf Optimizer", 2016 IEEE Student
Conference on Research and Development, 2016.

[14] Hossein Mombeini and AbdolrezaYazdani-

Chamzini, "Modeling Gold Price via Artificial
Neural Network", Journal of Economics,
Business and Management, 2015.
[15] ZurianiMustaffa and NurulAsyikin Zainal,
"A Literature Review On Gold Price Predictive
Techniques", 4th International Conference on

© 2022, IJSREM | www.ijsrem.com | Page 8

View publication stats