Chapter 1

Introduction
Introduction
Crude oil is gaining its important as a lifeline to the world’s economy in general and to some of
the nation’s economy in specific. Crude oil is the most actively traded commodity in the world.
The Indian oil and gas sector is one of the six core industries in India and plays a huge role in
global economy. As a developing country, it is committed to excel its economy in the upcoming
years. Increased oil prices have definite impact on world economy through employment, rising
inflation, decrease in dollar value all of which combine to economic slowdown. Any massive
increase or decrease in crude oil has its impact on the condition of stock markets throughout the
world. India is one of the biggest importers of crude oil in the world; ranks 3rd behind U.S and
China for import of crude oil. Oil accounts for about 30% of India’s total energy consumption.

The country’s total oil consumption is about 2.2 million barrels per day. India imports about
70% of its total oil consumption and it makes no exports. This accounts for one third of its total
imports. While falling crude oil, price is negative for oil-producing countries, India, being a net
importer, stands to benefit significantly. India’s macro-economic fundamentals such as inflation,
fiscal deficit and current account deficit (CAD) have improved meaningfully over the last couple
of years, a large part of which can be attributed to falling crude oil prices. The Indian oil and gas
sector is one of the six core industries in India and has very significant forward linkages with the
entire economy. The Indian oil and gas sector is of strategic importance and plays a
predominantly pivotal role in influencing decisions in all other spheres of the economy. Fall in
crude oil prices, therefore, helps the country save on import bill, thereby narrowing its CAD.
Price of crude oil is influenced by many factors like socio and political events, status of financial
markets. From medium to long run it is influenced by the fundamentals of demand and supply.
There is an emphasized need for wider and more intensive exploration for new finds, more
efficient and effective recovery, a more rational and optimally balanced global price regime- as
against the rather wide upward fluctuations of recent times, and a spirit of equitable common
benefit in global energy cooperation.

Decline in crude oil prices helps the government manage its finances better as it translates into
lower subsidies on petroleum products (LPG and kerosene), thereby resulting in lower fiscal
deficit. This, in turn, helps the government to remain committed to fiscal consolidation roadmap
without compromising on economic growth. The fall in global oil prices may be beneficial to
India, but it also has its downsides. Directly, it affects the exporters of petroleum producers in
the country. India is the sixth largest exporter of petroleum products in the world. This helps it
earn $60 billion annually. Any fall in oil prices negatively impacts exports. At a time when India
is running a trade deficit - high imports and low exports, any fall in exports is bad news.
Moreover, a lot of India’s trade partners and buyers of its exports are net oil exporters. A fall in
oil price may impact their economy, and hamper demand for Indian products. This would
indirectly affect India and its companies. Global oil fluctuations affect the economy of the
nations in a positive and negative manner depending whether the country is net importer or

1|Page
exporter of crude oil. India being a developing economy imports the 70% of its crude oil
requirements. For an investor and policy maker, it is imperative to understand the relationship
between crude oil prices and stock market. Volatile phase in the history of oil price shocks better
known by subprime crisis and concluded that international oil price variability has impact on
Indian stock market

What Is Crude Oil?

Crude oil is a naturally occurring petroleum product composed of hydrocarbon deposits and
other organic materials. A type of fossil fuel, crude oil is refined to produce usable products
including gasoline, diesel, and various other forms of petrochemicals. It is a nonrenewable
resource, which means that it can't be replaced naturally at the rate we consume it and is,
therefore, a limited resource.

Even though most crude oil is produced by a relatively small number of companies and often
located in remote locations far from the point of consumption, trading in crude oil on a global
basis has always been robust. Nearly 80% of international crude oil is transported through
waterways in large tankers and most of the rest by inland pipelines. OPEC controls 40 percent of
the world’s crude oil reserves and exports 55 percent of the oil traded internationally.

In oil trading, risk management techniques are extremely important for the various stakeholders
and participants, such as producers, exporters, marketers, processers and SME’s. Modern
techniques and strategies, including market- based risk management financial instruments like
crude oil futures, offered on the MCX platform can improve efficiencies and consolidate
competitiveness through price risk management.

Understanding Crude Oil

Crude oil is typically obtained through drilling, where it is usually found alongside other
resources, such as natural gas (which is lighter and therefore sits above the crude oil) and saline
water (which is denser and sinks below).
After its extraction, crude oil is refined and processed into a variety of forms, such as gasoline,
kerosene, and asphalt, for sale to consumers.
Although it is often called "black gold," crude oil has a range of viscosity and can vary in color
from black to yellow depending on its hydrocarbon composition. Distillation, the process by
which oil is heated and separated into different components, is the first stage in refining.
Although fossil fuels like coal have been harvested for centuries, crude oil was first discovered
and developed during the Industrial Revolution, and its industrial uses were developed in the
19th century. Newly invented machines revolutionized the way we do work, and they depended
on these resources to run.

Today, the world's economy is largely dependent on fossil fuels such as crude oil, and the
demand for these resources often sparks political unrest, as a small number of countries control
the largest reservoirs. Like any industry, supply and demand heavily affect the prices and

2|Page
profitability of crude oil. The United States, Saudi Arabia, and Russia are the leading producers
of oil in the world.
In the late 19th and early 20th centuries, the United States was one of the world's leading oil
producers, and U.S. companies developed the technology to make oil into useful products like
gasoline. During the middle and last decades of the 20th century, U.S. oil production fell
dramatically, and the U.S. became an energy importer.
Its major supplier was the Organization of the Petroleum Exporting Countries (OPEC), founded
in 1960, which consists of the world's largest (by volume) holders of crude oil and natural gas
reserves. As such, the OPEC nations had a great deal of economic leverage in determining
supply, and therefore the price, of oil in the late 1900s.
In the early 21st century, the development of new technology, particularly hydro-fracturing,
created a second U.S. energy boom, largely decreasing OPEC's importance and influence.
Heavy reliance on fossil fuels is cited as one of the main causes of global warming, a topic that
has gained traction in the past 20 years. Risks surrounding oil drilling include oil spills and ocean
acidification, which damage the ecosystem. In the 21st century, many manufacturers have begun
creating products that rely on alternative sources of energy, such as cars run by electricity, homes
powered by solar panels, and communities powered by wind turbines.
Factors influencing the market

• OPEC output or supply

• Changing scenarios in oil demand from emerging and developing countries

• Us crude and products inventories

• Refinery utilization rate

• Global geopolitics

• Speculative buying and selling

• Weather conditions
Supply

For several decades, the Organization of Petroleum Exporting Countries (OPEC) has been the
elephant on the world's trading floors, with its oil-producing member nations working together to
determine prices by boosting or reducing crude oil production. While OPEC's grip on the market
has loosened some in past years, its decisions continue to play a dominant role. OPEC's every
move is watched closely by governments, oil companies, speculators, hedgers, investors, traders,
policymakers, and consumers

OPEC's policies are affected, in turn, by geopolitical developments. Some of the world's top oil
producers are politically unstable or at odds with the West (issues pertaining to terrorism or
compliance with international laws, in particular, have been problematic). Some have faced

3|Page
sanctions by the U.S. and the United Nations. The supply crude oil is also determined by external
factors, which might include weather patterns, exploration and production (E&P) costs,
investments, and innovation.
In the past, supply disruptions triggered by political events have caused oil prices to shift
drastically; the Iranian revolution, Iran-Iraq war, Arab oil embargo, and Persian Gulf wars have
been especially notable. The Asian financial crisis and the global economic crisis of 2007-2008
also caused fluctuations.
Demand
Strong economic growth and industrial production tend to boost the demand for oil—as reflected
in changing demand patterns by non-OECD nations, which have grown rapidly in recent years
Other important factors that affect demand for oil include transportation (both commercial and
personal), population growth, and seasonal changes. For instance, oil use increases during busy
summer travel seasons and in the winters, when more heating fuel is consumed.

Investing in Oil

Investors may purchase two types of oil contracts: futures contracts and spot contracts. To the
individual investor, oil can be a speculative asset, a portfolio diversifier, or a hedge against
related positions.
Spot Contracts
The price of the spot contract reflects the current market price for oil, whereas the futures price
reflects the price buyers are willing to pay for oil on a delivery date set at some point in the
future.
The futures price is no guarantee that oil will actually hit that price in the current market when
that date comes. It is just the price that, at the time of the contract, purchasers of oil are
anticipating. The actual price of oil on that date depends on many factors.

Most commodity contracts that are bought and sold on the spot markets take effect immediately:
Money is exchanged, and the purchaser accepts delivery of the goods. In the case of oil, the
demand for immediate delivery versus future delivery is small, in no small part due to the
logistics of transporting oil.
Investors, of course, don't intend to take delivery of commodities at all (although there have been
cases of investor errors that have resulted in unexpected deliveries), so futures contracts are more
commonly used by traders and investors.

Futures Contracts

An oil futures contract is an agreement to buy or sell a certain number of barrels of oil at a
predetermined price, on a predetermined date. When futures are purchased, a contract is signed

4|Page
between buyer and seller and secured with a margin payment that covers a percentage of the total
value of the contract.
End-users of oil purchase on the futures market in order to lock in a price; investors buy futures
essentially as a gamble on what the price will actually be down the road, and they profit if they
guess correctly. Typically, they will liquidate or roll over their futures holdings before they
would have to take delivery.
There are two major oil contracts that are closely watched by oil market participants. In North
America, the benchmark for oil futures is West Texas Intermediate (WTI) crude, which trades on
the New York Mercantile Exchange (NYMEX). In Europe, Africa, and the Middle East, the
benchmark is North Sea Brent Crude, which trades on the Intercontinental Exchange (ICE).
While the two contracts move somewhat in unison, WTI is more sensitive to American economic
developments, and Brent responds more to those overseas.
While there are multiple futures contracts open at once, most trading revolves around the front-
month contract (the nearest futures contract). For this reason, it's is known as the most
active contract.
Spot vs. Future Oil Prices

Futures prices for crude oil can be higher, lower or equal to spot prices. The price difference
between the spot market and the futures market says something about the overall state of the oil
market and expectations for it. If the futures prices are higher than the spot prices, this usually
means that purchasers anticipate the market will improve, so they are willing to pay
a premium for oil to be delivered at a future date. If the futures prices are lower than the spot
prices, this means that buyers expect the market to deteriorate.
"Backwardation" and "contango" are two terms used to describe the relationship between
expected future spot prices and actual futures prices. When a market is in contango, the futures
price is above the expected spot price. When a market is in normal backwardation, the futures
price is below the expected future spot price. The prices of different futures contracts can also
vary depending on their projected delivery dates.

Forecasting Oil Prices

Economists and experts are hard-pressed to predict the path of crude oil prices, which are volatile
and depend on many variables. They use a range of forecasting tools and depend on time to
confirm or disprove their predictions. The five models used most often are:

• Oil futures prices: Central banks and the International Monetary Fund (IMF) mainly use
oil futures contract prices as their gauge. Traders in crude oil futures set prices by two
factors: supply and demand and market sentiment. However, futures prices can be a poor
predictor, because they tend to add too much variance to the current price of oil.

5|Page
• Regression-based structural models: Statistical computer programming calculates the
probabilities of certain behaviors on the price of oil. For instance, mathematicians may
consider forces such as events in OPEC member nations, inventory levels, production
costs, or consumption levels. Regression-based models have strong predictive power, but
their creators may fail to include one or more factors, or unexpected variables may step in
to cause these regression-based models to fail.

• Time-series analysis: Some economists use time-series models, such as exponential

smoothing models and autoregressive models, which include the categories
of ARIMA and the ARCH/GARCH, to correct for the limitations of oil futures prices.
These models analyze the history of oil at various points in time to extract meaningful
statistics and predict future values based on previously observed values. Time-series
analysis sometimes errs, but usually produces more accurate results when economists
apply it to shorter time spans.

• Bayesian autoregressive models: One way to improve upon the standard regression-based
model is by adding calculations to gauge the probability of the impact of certain predicted
events on oil. Most contemporary economists like to use the Bayesian vector
autoregressive (BVAR) model for predicting oil prices. A 2015 International Monetary
Fund working paper noted that these models work best when used on a maximum 18-
month horizon and when a smaller number of predictive variables are inserted. BVAR
models accurately predicted the price of oil during the years 2008-2009 and 2014-2015.

• Dynamic stochastic general equilibrium graphs: Dynamic stochastic general equilibrium

(DSGE) models use macroeconomic principles to explain complex economic
phenomena; in this case, oil prices. DSGE models sometimes work, but their success
depends on events and policies remaining unchanged, as DSGE calculations are based on
historical observations.

6|Page
Review of Literature

J. Jaya Selvi, R. Kaviya Shree, J. Krishnan (2018) analyzed crude oil prices using Arima
model. In order to forecast the crude oil price for future periods the time series analysis is carried
out in time series plot and ARIMA model and the future values are forecasted using the model.
The prices of crude oil are normal during 1946 – 1972, which was identified as “The early
postwar era”. Then afterwards the prices of crude oil were increasing due to major oil crisis, this
occurred in 1973. The prices of crude oil in the upcoming year will in an increasing manner.
Even though the use of oil is different, the prices of oil should be maintained or at some instance,
it will lead to a huge problem to the economy of a country. Hence, the prices should be
normalized and special care should be given in monitoring the prices of oil.

Jessin Shah P, Dr. G Kiruthiga (2020) Crude oil is one of the most important chemical and energy
resources. The crude oil and its price affect the economic and social activities, therefore crude effective
crude oil forecast can help stabilize economic development and prevent energy crisis. The WTI crude oil
data from the period of 1987 to 2020 were analyzed and showed the non stationarity of data. We
conducted data analysis and preprocessing prepared for ARIMA. To value the commodity, three crude oil
benchmarks are set; they are Brent, West Texas Intermediate (WTI) and Dubai/Oman oil benchmarks.
WTI crude oil price data refers to crude oil extracted in the U.S. is used in this study.

Day(2006) compared the volatility of crude oil since November 2003 to March 2006 in the fitted
GARCH (1,1) and EGARCH (1,1) model, using the least squares method to estimate the parameters and
make predictions [18]. The results show that GARCH / EGARCH model estimated volatility parameters
are in line with the volatility of volatility data for information response. However, with the implied
volatility and GARCH model to estimate the parameters of the integrated volatility might get better
prediction results.

S. kulkarni and Imad Haider (2009) forecasted crude oil prices on the short-term. In addition, we tested
the relation between crude oil futures prices and spot price, and if futures are good predictors to the spot
applying nonlinear ANN model. Namely, Daily spot price for WTI and futures prices for 1, 2, 3, and 4,
months to maturity was considered. Data was obtained from Energy Information Administration covering
the period from 1996 to 2007. Although, it could be argued that the relation between spot and futures
could be different during the day. In other words, testing with intraday data could produce different
results. However, intraday data for crude oil prices is not available. Finally, our future research continue
to investigate other variable which could lead to improving the short-term forecast, such as heating oil
prices, interest rate, and gold prices.

7|Page
Auto Regressive Integrated Moving Average (ARIMA) model

An autoregressive integrated moving average, or ARIMA, is a statistical analysis model that

uses time series data to either better understand the data set or to predict future trends.
A statistical model is autoregressive if it predicts future values based on past values. For
example, an ARIMA model might seek to predict a stock's future prices based on its past
performance or forecast a company's earnings based on past periods.

An autoregressive integrated moving average model is a form of regression analysis that gauges
the strength of one dependent variable relative to other changing variables. The model's goal is to
predict future securities or financial market moves by examining the differences between values
in the series instead of through actual values.

Auto regression (AR) refers to a model that shows a changing variable that regresses on its own
lagged or prior, values.

Integrated (I) represents the differencing of raw observations to allow for the time series to
become stationary (i.e., data values are replaced by the difference between the data values and
the previous values).

Moving average (MA) incorporates the dependency between an observation and a residual error
from a moving average model applied to lagged observations.

• Autoregressive integrated moving average (ARIMA) models predict future values based
on past values.
• ARIMA makes use of lagged moving averages to smooth time series data.
• They are widely used in technical analysis to forecast future security prices.
• Autoregressive models implicitly assume that the future will resemble the past.
• Therefore, they can prove inaccurate under certain market conditions, such as financial
crises or periods of rapid technological change.

ARIMA Parameters

Each component in ARIMA functions as a parameter with a standard notation. For ARIMA
models, a standard notation would be ARIMA with p, d, and q, where integer values substitute
for the parameters to indicate the type of ARIMA model used. The parameters can be defined as:

• p: the number of lag observations in the model; also known as the lag order.
• d: the number of times that the raw observations are differenced; also known as the
degree of differencing.
• q: the size of the moving average window; also known as the order of the moving
average.

8|Page
Autoregressive Integrated Moving Average (ARIMA) and Stationarity

In an autoregressive integrated moving average model, the data are differenced in order to make
it stationary. A model that shows stationarity is one that shows there is constancy to the data over
time. Most economic and market data show trends, so the purpose of differencing is to remove
any trends or seasonal structures.
Seasonality, or when data show regular and predictable patterns that repeat over a calendar year,
could negatively affect the regression model. If a trend appears and stationarity is not evident,
many of the computations throughout the process cannot be made with great efficacy.

The ARIMA model predicts a given time series based on its own past values. It can be used for
any non-seasonal series of numbers that exhibits patterns and is not a series of random events.
For example, sales data from a clothing store would be a time series because it was collected
over a period of time. One of the key characteristics is the data is collected over a series of
constant, regular intervals. A modified version can be created to model predictions over multiple
seasons.

For a period of multiple seasons, the data must be corrected to account for differences between
the seasons. For example, holidays fall on different days of the year, causing a seasonal effect to
the data. Sales may be artificially higher or lower depending on where the holiday falls in the
calendar. The data scientist must be able to seasonally adjust the data to provide an accurate
prediction for future sales.

The ARIMA model is becoming a popular tool for data scientists to employ for forecasting
future demand, such as sales forecasts, manufacturing plans or stock prices. In forecasting stock
prices, for example, the model reflects the differences between the values in a series rather than
measuring the actual values.

Unit root test

What is “Unit Root”

A unit root (also called a unit root process or a difference stationary process) is
a stochastic trend in a time series, sometimes called a “random walk with drift”; If a time series
has a unit root, it shows a systematic pattern that is unpredictable.

What is a Unit Root Test?

Unit root tests are tests for stationarity in a time series. A time series has stationarity if a shift in
time doesn’t cause a change in the shape of the distribution; unit roots are one cause for non-
stationarity.
These tests are known for having low statistical power. Many tests exist, in part, because none
stand out as having the most power.

9|Page
• The Dickey Fuller Test (sometimes called a Dickey Pantula test), which is based on
linear regression. Serial correlation can be an issue, in which case the Augmented Dickey-
Fuller (ADF) test can be used. The ADF handles bigger, more complex models. It does have
the downside of a fairly high Type I error rate.

10 | P a g e
 Chapter 2
Methodology of the Study
Research problem statement

“A study on analyzing and Forecasting Crude oil prices in India using ARIMA model”

Research Objectives

• The main objective is to study and analyze the crude oil prices.
• To forecast the crude oil prices using time series analysis

Research Design

The report used descriptive research design because here the report include the study about
forecasting the crude oil prices in India

Sample size

The report used data of crude oil prices of 5 years starting from July 2016 to May 2021

Data collection

There are two types of data collection techniques

1) Primary data
2) Secondary data

Here the report use secondary sources of data

Secondary data are those data which are collected by researcher and which are used by another
researcher for his present study. Like websites of commodities, journals, magazines, research
papers etc.

Data analysis tools and techniques

In the report tables, charts and ARIMA model of time series analysis is used to study and
forecast the crude oil prices in India.

11 | P a g e
Limitations

• Accuracy of available data

• Knowledge constraint
• Sometimes there is error contains in analyzing and forecasting prices through time series
analysis

12 | P a g e
 Chapter 3
Data analysis and Interpretation
Data Analysis

Crude oil price graph

1st crude oil price

800

400

-400

-800

-1,200

-1,600
III IV I II III IV I II III IV I II III IV I II III IV I II
2016 2017 2018 2019 2020 2021

Interpretation

The above chart indicates the graphical representation of the crude oil prices during the period of
July 2016 to May 2021. It shows that there is upward and downward moments in the prices of
crude oil. So it indicates that series is not stationary.

13 | P a g e
Descriptive statistics

_1ST_CRU...
Mean 32.98155
Median 87.15000
Maximum 695.5300
Minimum -1418.800
Std. Dev. 391.4553
Skewness -1.469597
Kurtosis 6.193728

Jarque-Bera 45.52700
Probability 0.000000

Sum 1912.930
Sum Sq. Dev. 8734525.

Observations 58

Interpretation

The above table represents the descriptive analysis of crude oil prices traded in India. Mean
value of the crude oil is 32.9815, median value is 87.1500, and standard deviation is 391.4553

14 | P a g e
Correlogram of Crude oil prices with 1st level difference

Date: 01/01/09 Time: 00:19

Sample (adjusted): 2016M09 2021M05
Included observations: 57 after adjustments
Autocorrelation Partial Correlation AC PAC Q-Stat Prob

1 -0.102 -0.102 0.6259 0.429

2 -0.287 -0.300 5.6575 0.059
3 -0.233 -0.335 9.0308 0.029
4 -0.003 -0.238 9.0312 0.060
5 0.016 -0.292 9.0478 0.107
6 0.135 -0.152 10.241 0.115
7 0.050 -0.143 10.412 0.166
8 -0.121 -0.272 11.415 0.179
9 0.151 0.075 13.015 0.162
10 -0.028 -0.040 13.070 0.220
11 -0.042 0.038 13.200 0.280
12 -0.102 -0.012 13.982 0.302
13 -0.005 -0.049 13.984 0.375
14 -0.124 -0.272 15.179 0.366
15 0.162 -0.150 17.292 0.302
16 0.224 -0.013 21.410 0.163
17 0.017 0.063 21.434 0.207
18 -0.284 -0.199 28.406 0.056
19 -0.054 -0.035 28.668 0.071
20 0.117 0.073 29.902 0.071
21 0.070 0.085 30.359 0.085
22 0.057 0.137 30.670 0.103
23 -0.205 -0.109 34.829 0.054
24 0.066 0.101 35.277 0.064

Interpretation

The Correlogram of crude oil price with 1st level difference shows that spikes except AR (2), MA
(2), AR (3), and AR(5) are under the dotted lines. So this shows that series is not consider being
stationary series it is a non-stationary series.

15 | P a g e
Unit root test of crude oil price at level

Null Hypothesis: D(_1ST_CRUDE_OIL_PRICE) has a unit root

Exogenous: Constant
Lag Length: 2 (Automatic - based on SIC, maxlag=10)

t-Statistic Prob.*

Augmented Dickey-Fuller test statistic -7.178585 0.0000

Test critical values: 1% level -3.557472
5% level -2.916566
10% level -2.596116

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(_1ST_CRUDE_OIL_PRICE,2)
Method: Least Squares
Date: 01/01/09 Time: 00:12
Sample (adjusted): 2016M12 2021M05
Included observations: 54 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

D(_1ST_CRUDE_OIL_PRICE(-1)) -1.900673 0.264770 -7.178585 0.0000

D(_1ST_CRUDE_OIL_PRICE(-1),2) 0.681745 0.194816 3.499435 0.0010
D(_1ST_CRUDE_OIL_PRICE(-2),2) 0.334672 0.131636 2.542401 0.0142
C 11.26219 56.58919 0.199017 0.8431

R-squared 0.638802 Mean dependent var 12.49759

Adjusted R-squared 0.617130 S.D. dependent var 671.8924
S.E. of regression 415.7435 Akaike info criterion 14.96920
Sum squared resid 8642134. Schwarz criterion 15.11653
Log likelihood -400.1684 Hannan-Quinn criter. 15.02602
F-statistic 29.47604 Durbin-Watson stat 2.022512
Prob(F-statistic) 0.000000

Interpretation

To check the stationarity of data the Augmented Dickey-Fuller test of unit root test statistics is
applied. Here to make series stationary crude oil price with level and intercept is applied. The
probability value is important which should be less than 0.05 or 5% and in the above table we
can see that probability value is less than 0.05 i.e. value is 0.0000. So we can reject null
hypothesis and can say that the series is stationary.

16 | P a g e
ARIMA method

Dependent Variable: _1ST_CRUDE_OIL_PRICE

Method: ARMA Maximum Likelihood (OPG - BHHH)
Date: 08/26/21 Time: 22:39
Sample: 2016M08 2021M05
Included observations: 58
Convergence not achieved after 500 iterations
Coefficient covariance computed using outer product of gradients

Variable Coefficient Std. Error t-Statistic Prob.

C 35.14253 60.06108 0.585113 0.5613

AR(1) 1.113103 0.316466 3.517292 0.0010
AR(2) -1.100923 0.481686 -2.285562 0.0269
AR(3) -0.618181 0.577900 -1.069701 0.2903
AR(4) 0.793055 0.478563 1.657158 0.1043
AR(5) -0.841693 0.195003 -4.316306 0.0001
MA(1) -0.991914 68.10707 -0.014564 0.9884
MA(2) 0.990328 131.3333 0.007541 0.9940
MA(3) 0.997545 234.2606 0.004258 0.9966
MA(4) -0.991544 191.5162 -0.005177 0.9959
MA(5) 0.992590 164.9407 0.006018 0.9952
SIGMASQ 75867.71 9746356. 0.007784 0.9938

R-squared 0.496215 Mean dependent var 32.98155

Adjusted R-squared 0.375744 S.D. dependent var 391.4553
S.E. of regression 309.2884 Akaike info criterion 14.77732
Sum squared resid 4400327. Schwarz criterion 15.20362
Log likelihood -416.5422 Hannan-Quinn criter. 14.94337
F-statistic 4.118973 Durbin-Watson stat 1.758390
Prob(F-statistic) 0.000313

Inverted AR Roots .63-.76i .63+.76i .38+.92i .38-.92i

-.89
Inverted MA Roots .52-.85i .52+.85i .47+.88i .47-.88i
-1.00

17 | P a g e
Correlogram of residuals squared
Date: 01/01/09 Time: 00:59
Sample (adjusted): 2016M08 2021M05
Q-statistic probabilities adjusted for 10 ARMA terms

Autocorrelation Partial Correlation AC PAC Q-Stat Prob

1 0.107 0.107 0.6978

2 -0.049 -0.061 0.8481
3 -0.040 -0.028 0.9501
4 0.105 0.112 1.6654
5 -0.058 -0.088 1.8844
6 0.057 0.087 2.1035
7 -0.007 -0.024 2.1067
8 -0.051 -0.061 2.2850
9 0.013 0.052 2.2969
10 -0.145 -0.193 3.8189
11 -0.147 -0.097 5.4215 0.020
12 -0.008 0.018 5.4263 0.066
13 -0.016 -0.074 5.4467 0.142
14 -0.149 -0.104 7.2065 0.125
15 0.054 0.089 7.4383 0.190
16 0.215 0.195 11.271 0.080
17 0.080 0.065 11.821 0.107
18 -0.157 -0.162 13.958 0.083
19 -0.098 -0.085 14.816 0.096
20 0.034 0.022 14.923 0.135
21 -0.022 -0.113 14.966 0.184
22 -0.033 -0.059 15.068 0.238
23 -0.224 -0.267 20.072 0.093
24 -0.015 0.007 20.095 0.127

Interpretation

After using ARIMA model (1,1,5) we can say that all the spikes in the Correlogram are within
the lines which indicates all the information are captured and we conclude that adjusted Arima
model is most ideal so the forecast will be based on this model.

18 | P a g e
2,000
1,500
1,000
500
0
-500
-1,000
-1,500
-2,000
M6 M7 M8 M9 M10 M11 M12
2021

_1ST_CRUDEF ± 2 S.E.

Interpretation

From the chart we can see that there is forecasting period is June 2021 to December 2021 this
forecasting is based on final Arima model.

19 | P a g e
 800

400

-400

-800

-1,200

-1,600
III IV I II III IV I II III IV I II III IV I II III IV I II III IV
2016 2017 2018 2019 2020 2021

_1ST_CRUDEF 1st crude oil price

Interpretation

This graph indicates the actual and forecasted series of the data where red line indicates actual
series of data and blue line indicates the forecasted series. So we can forecast that in the 1 st and
2nd quarter there is decline in the prices and in 3rd quarter there is increasing in the prices and
again in 4th quarter of 2021 there is decline in the prices of crude oil.

20 | P a g e
 Chapter 4
Findings and conclusion
Findings and conclusion

ARIMA is one of the useful techniques in forecasting the data. The results are summarized using
ARIMA model of time series analysis and future values are forecasted using this model. After
using Correlogram and unit root series to check stationarity of the data we arrive at final Arima
model which is used for forecasting of future prices. With this model forecasting for the June
2021 to December 2021 is done. In the 1 st and 2nd quarter there is decline in the prices and in 3rd
quarter there is increasing in the prices and again in 4 th quarter of 2021 there is decline in the
prices of crude oil.

ARIMA model is appropriate, the static model to better simulate historical data but it does not
consider the impact of external factors also the historical data selected time period is shorter.

21 | P a g e
Bibliography

https://www.investopedia.com/terms/c/crude-oil.asp

https://www.mcxindia.com/products/energy/crude-oil

https://www.investopedia.com/terms/a/autoregressive-integrated-moving-average-arima.asp

https://www.statisticshowto.com/unit-root/

https://www.investopedia.com/articles/investing/072515/top-factors-reports-affect-price-oil.asp

22 | P a g e
Annexure

23 | P a g e
Month price

Jul-16 2,966.28
Aug-16 3,004.16
Sep-16 3,006.05
Oct-16 3,290.46
Nov-16 3,056.29
Dec-16 3,572.84
Jan-17 3,649.89
Feb-17 3,647.03
Mar-17 3,355.09
Apr-17 3,365.04
May-17 3,213.83
Jun-17 2,975.01
Jul-17 3,071.23
Aug-17 3,194.46
Sep-17 3,413.19
Oct-17 3,574.53
Nov-17 3,887.81
Dec-17 3,930.99
Jan-18 4,215.16
Feb-18 4,085.16
Mar-18 4,171.72
Apr-18 4,516.93
May-18 4,959.75
Jun-18 4,879.75
Jul-18 4,992.51
Aug-18 4,942.53
Sep-18 5,448.55
Oct-18 5,648.69
Nov-18 4,476.09
Dec-18 3,822.07
Jan-19 4,003.08
Feb-19 4,352.94
Mar-19 4,432.04
Apr-19 4,761.33
May-19 4,664.02
Jun-19 4,149.67
Jul-19 4,230.22
Aug-19 4,102.97

2
4
Sep-19 4,282.87
Oct-19 4,069.14
Nov-19 4,314.32
Dec-19 4,509.77
Jan-20 4,395.91
Feb-20 3,811.78
Mar-20 2,392.98
Apr-20 1,603.02
May-20 2,298.55
Jun-20 2,987.46
Jul-20 3,156.01
Aug-20 3,243.75
Sep-20 2,984.05
Oct-20 2,931.66
Nov-20 3,141.53
Dec-20 3,589.35
Jan-21 3,918.56
Feb-21 4,399.41
Mar-21 4,646.36
Apr-21 4,684.64
May-21 4,879.21