ESTIMATION OF CRUDE OIL PRICES BY THE

APPLICATION OF
SVR AND THE GRID SEARCH CROSS VALIDATION
ALGORITHM

Name: P.Bharath Kumar

Roll Number: 23N31D5812
 ABSTRACT:
• Crude oil is the primary fuel for most vehicles on Earth, and fluctuations in its price have far-reaching
consequences for ecosystems, economies, and oil exploration and production. Any entity, from businesses
to governments, may benefit greatly from accurate predictions of oil prices.
• The extreme volatility of the commodity makes oil price forecasting notoriously difficult, despite the existence
of several devised methodologies for the purpose. Numerous sectors, including economics and science, use
forecasting models to make informed decisions based on projected future occurrences. If you want to know
what the world's environment is going to be like when the oil price drops, you need a flawless prediction.
• It will show you the consequences of doing nothing and how much control you have over the future. The
gasoline expenses have gone down because oil prices have gone down. Consequently, there would
undoubtedly be an increase in carbon emissions due to customers' increased oil use. Furthermore,
discouraging research and development of clean and renewable energy sources would occur. Conversely, a
decline in worldwide oil and gas exploration and exploitation might result from persistently low oil prices.
• One major factor influencing world economies is the volatility of oil prices. Worldwide
economic activity would get a little lift from a decline in oil prices, but oil industry owners
would see a decline in their revenue.
• According to new data from the World Bank, the global gross domestic product (GDP)
would rise by 0.5% for every 30% drop in oil prices. Inflation would decline as a result of
lower pricing for goods and services brought about by a decline in oil prices. As a result, if
an issue does arise, there is a possibility of making an accurate and approximate
prognosis to address it.
 EXISTING SYSTEM:

• The prediction of the crude oil rates based on the previous datasets on the data
and prices as the feature _list are inputs and target list are predicted values.

• The implementation was on the Linear Regression Model which is feasible to some
extend for the prediction of the crude oil prices.

• The implementation is on predicting the crude oil prices for the days using Linear
Regression Python Machine Algorithm and plotting the graph based on the
prediction.
 DISADVANTAGES:

• Using Linear Regression algorithm gives less approximate prediction compared to

SVR Algorithm in the proposed model in the project.

• As well the feature list and target list fitted into the algorithm gives less predicting
prices compared to the SVR, Comparatively Linear regression performs poorly when
there are non-linear relationships.

• They are not naturally flexible enough to capture more complex patterns, and
adding the right interaction terms or polynomials can be tricky and time-consuming.
 PROPOSED SYSTEM:
 We have implemented SVR algorithm (Support Vector Regression) of Machine learning using Python. The predictions are

most approximate with SVR Algorithms as they Linear or Gaussian. The algorithm automatically uses the kernel function

that is most appropriate to the data.

 SVM uses the linear kernel when there are many attributes (more than 100) in the training data, otherwise it uses the

Gaussian kernel. In the proposed system we have taken the datasets which has the price and days based on the dataset

we have made feature list and target list where the target list is price values and feature list is the days.

 After the analysis of data is done we have fitted both feature list and target list using Python Machine learning SVM

Algorithm and predicted the values for 1,30 and 365 days from the last day of the dataset values. Finally, we have

plotted a graph based on the results from the predicted analysis done with SVM Algorithm.
 ADVANTAGES:
• SVMs are a new promising non-linear, non-parametric classification technique, which already showed good results in
the medical diagnostics, optical character recognition, electric load forecasting and other fields. Applied to solvency
analysis, the common objective of all these, it has a regularization parameter, which makes the user think about
avoiding over-fitting.

• Secondly it uses the kernel trick, so you can build in expert knowledge about the problem via engineering the kernel.
Thirdly an SVM is defined by a convex optimization problem (no local minima) for which there is efficient methods
(e.g. SMO). Lastly, it is an approximation to a bound on the test error rate, and there is a substantial body of theory
behind it which suggests it should be a good idea.

• The results that are generated by this algorithm gives more approximate and accurate calculations of the price
prediction value compared to the other prediction algorithm for the dataset provided.
SYSTEM ARCHITECTURE DIAGRAM:
 SYSTEM SPECIFICATION:
HARDWARE REQUIREMENTS:
• Processor: Intel i3
• RAM: 4GB
• Hard Disk :1TB

SOFTWARE REQUIREMENTS:
• Operating System: Windows 10.
• IDE: PyCharm
• Coding Language: Python.
• Front-End: Python.
• Designing: HTML, CSS, JAVASCRIPT.
• Data Base: MySQL (WAMP Server).
 LITERATURE SURVEY:
"Forecasting copper prices using hybrid adaptive neuro-fuzzy inference system and genetic
algorithms“
Authors: Z Alameer, MA Elaziz, et al.

An accurate forecasting model for the price volatility of minerals plays a vital role in future investments and

decisions for mining projects and related companies. In this paper, a hybrid model is proposed to provide an

accurate model for forecasting the volatility of copper prices. The proposed model combines the adaptive neuro-

fuzzy inference system (ANFIS) and genetic algorithm (GA). Genetic algorithms are used for estimating the ANFIS

model parameters. The results of the proposed model are compared to other models, including ANFIS, support

vector machine (SVM), generalized autoregressive conditional heteroscedasticity (GARCH), and autoregressive

integrated moving average (ARIMA) models. The empirical results confirm the superiority of the hybrid GA–ANFIS

model over other models. The proposed model also improves the forecasting accuracy obtained from the ANFIS,

SVM, GARCH, and ARIMA models by a 62.92%, 36.38%, 91.72%, and 42.19% decrease in mean square error,
"An adaptive forecasting approach for copper price volatility through hybrid and non-
hybrid models"
Authors: D García, W Kristjanpoller et al.
This article studies monthly volatility forecasting for the copper market, which is of
practical interest for various participants such as producers, consumers, governments, and
investors. we propose a framework composed of a set of time series models such as Auto-
Regressive Integrated Moving Average (ARIMA) and Generalized Auto-Regressive Conditional
Heteroscedasticity (GARCH), non-parametric models from soft computing, e.g. Artificial Neural
Networks (ANN) and Fuzzy Inference Systems (FIS), and hybrid specifications of both. The
adaptability characteristic of these models in exogenous variables, their configuration
parameters and window size, simultaneously, are provided by a Genetic Algorithm in pursuit of
achieving the best possible forecasts. Also, recognized drivers of this specific market are
considered.
"Financial time series forecasting using support vector machines"
Authors: K Kim and G. Elsevier et al.
Support vector machines (SVMs) are promising methods for the prediction of
financial time-series because they use a risk function consisting of the empirical error and a
regularized term which is derived from the structural risk minimization principle. This study
applies SVM to predicting the stock price index. In addition, this study examines the feasibility
of applying SVM in financial forecasting by comparing it with back-propagation neural
networks and case-based reasoning. The experimental results show that SVM provides a
promising alternative to stock market prediction.
 MODULES:

1.ADD RESOURCE

2.ALLOCATE RESOURCE

3.USER QUERIES

4.GRAPH ANALYSIS
1.ADD RESOURCE
The resources have been uploading in database in order to view for users. Resources can be
uploading, modify or delete. The added resources can be visible to user and user can apply with
required details. User can add details with their quotation of application. So that user can implement
their work.

2.ALLOCATE RESOURCE
The received applications are viewed by admin. Admin then, find the available space and view
application and based on that, algorithm applied to sort the best user to be allocated or the allocation
space will be measured according to the quotation of user had submitted. The allocated resource can
be utilized by user and admin need to intimate the user that how much resource have been allocated.
3.USER QUERIES
Users can have queries about the process. This part of project is dedicated to make
and get response for queries that are needed to answerable. The major part of the modules is
making project as interactive one, queries have been very normally arise to users regarding
different details about the process.

4.GRAPH ANALYSIS
Graph analysis is the part where admin can knows the statistics about process of
details. The data are taken from the project flow and it shows until updated value. The data
are gives clear solution to admin that part of improvement and user satisfaction and other
factors.
 SYSTEM DESIGN:
COMPONENT DIAGRAM:
a.User
COMPONENT DIAGRAM:
b.Admin
ER DIAGRAM:
a. User
ER DIAGRAM:
b. Admin
USE CASE DIAGRAM:
a. User
USE CASE DIAGRAM:
b. Admin
CLASS DIAGRAM:
DATA FLOW DIAGRAM:
a. User
DATA FLOW DIAGRAM:
b. Admin
ACTIVITY DIAGRAM:
a. User
ACTIVITY DIAGRAM:
b.Admin
SEQUENCE DIAGRAM:
a. User
SEQUENCE DIAGRAM:
b. Admin