1

CHAPTER ONE
1.0 INTRODUCTION
Natural gas is a naturally occurring hydrocarbon gas mixture consisting primarily of methane,
with other hydrocarbons, carbon dioxide, nitrogen and hydrogen sulfide. Natural gas is an
important energy source to provide heating and electricity. It is also used as fuel for vehicles and
as a chemical feedstock in the manufacture of plastics and other commercially important organic
chemicals.
Natural gas is found in deep underground natural rock formations or associated with other
hydrocarbon reservoirs in coal beds and as methane clath-rates. Petroleum is also another
resource found in proximity to and with natural gas. Most natural gas was created over time by
two mechanisms: biogenic and thermo-genic. Biogenic gas is created by methano-
genic organisms in marshes, bogs, landfills, and shallow sediments. Deeper in the earth, at
greater temperature and pressure, thermo-genic gas is created from buried organic material.
Before natural gas can be used as a fuel, it must undergo processing to clean the gas and remove
impurities, including water, to meet the specifications of marketable natural gas. The by-products
of processing include ethane, propane, butanes, pentanes, and higher molecular
weight hydrocarbons, hydrogen sulfide (which may be converted into pure sulfur), carbon
dioxide, water vapor, and sometimes helium and nitrogen.
Natural gas is often informally referred to simply as gas, especially when compared to other
energy sources such as oil or coal.

1.2 SCOPE AND LIMITATION
1.2.1 SCOPE
The study considers natural gas variables of the Nigerian Liquidified Natural Gas (NLNG).
1.2.2 LIMITATION
Research is an area where Nigeria as a Country is lagging behind. Most times the reason is not
totally that Nigerian students are lazy rather the fact is that effort made in getting relevant
information in most cases proves abortive. This is one of the challenges and limitations we had
in carrying out the research. As Statisticians, without a well-presented data, the overall result will
be highly misleading how much more when there is no data. Also, sourcing for necessary
information was really tasking. Too many factors militated against the quick completion of this
research but the afore-mentioned are the predominant few.
2

1.3 AIMS AND OBJECTIVES
To develop a time series model for the data.
To forecast natural gas variables using the data.
1.4 LITERATURE REVIEW
Some of the greatest contributors to the theme of the research includes:
Herbert (1986) in an article on data analysis of sales of natural gas performed a Regression
diagnostics and a time series analysis of residuals are used to help define regression equations
and to identify, the weaknesses of these equations for explaining monthly deliveries of natural
gas to residential customers in the United States for the time period April 1979 through March
1983. More than 99% of the monthly variation in deliveries is explained by a linear regression
equation which includes heating degree days, cooling degree days, and the price of natural gas as
independent variables. Final estimated relationships yield useful monthly and annual estimates of
natural gas deliveries to residential customers in the United States for the time period April 1983
through March 1984. Most importantly, the estimated results when used in conjunction with the
diagnostics and the time series analysis of the residuals indicate the possible strengths,
weaknesses, and applicability of the estimated relationships
Hill et al (1994) in an article on petroleum production systems attempts to maximize production
in two connected systems, the reservoir and the artificial structures related to a well field.
Determining whether it is economically feasible to drill a discovered source requires a familiarity
with a variety of topics and theories. This book devotes the first six chapters to these concepts,
including production from under-saturated oil reserves, production from two-phase reservoirs,
production from natural gas reservoirs, skin effects, and gravel pack completions. The second
part of the production engineer`s mission is maximization of output in the artificial structures,
which include the well, bottom-hole, and wellhead assemblies, and the surface gathering,
separation, and storage facilities. This general system is discussed throughout the remainder of
the book. Chapter topics include wellbore flow performance, forecast of well production, well
test design and data acquisition, matrix acidizing, hydraulic fracturing for well simulation, gas
lift, and systems analysis.
Sundet et al (1994) in a journal on evaluation of natural gas made a comparative study of nine
synthesis gas technologies have been carried out. The basis was syngas feed to a 2500 MTPD
methanol plant. Material and energy balances were calculated, investments and operational costs
were estimated, and a comparison was made. The comparison indicates that auto-thermal
reforming and combined reforming with pre-reformer are preferable among the proven
technologies. Among emerging technologies GHR, KRES and CPO have an economic
advantage. However, the margins are probably too small to be significant.
3

Munoz and Sailor (1997) in a write-up on sensitivity of electricity and natural gas consumption
developed a method for assessing the sensitivity of electricity and natural gas consumption to
climate at regional scales. The approach involves a multiple-regression analysis of historical
energy and climate data, and has been applied to eight of the most energy-intensive states,
representing 42% of the total annual energy consumption in the United States. Statistical models
were developed using two sets of independent variablesprimitive variables such as
temperature, relative humidity, and wind speed, and derived variables including cooling degree
days, heating degree days, and enthalpy latent days. The advantages and disadvantages of both
modeling approaches are discussed in this paper, along with sample results for a combined
analysis of residential and commercial consumption in eight states.
Radetzki (1999) Market forces on natural gas analyzed the emergent commercial forces that
promote increasing competition in the European natural gas market. The paper begins by
describing the traditional market organization, along with its monopolistic elements and
inflexibilities. It goes on to illuminate the destabilizing frustration among producers with fast
growing supply potential, caused by the limited growth in demand under prevailing market
arrangements. Two emergent commercial forces promoting competition are then dealt with.
The first is the increasingly widespread effort by large consumers to procure gas on improved
terms by circumventing the national transmission companies. The important role played by
Wingas in this respect is reviewed in detail. The second is the impending impact of the
Interconnector, a gas pipe between the UK and the continent, which becomes operational during
1998. The paper argues that these commercial developments are undermining existing
monopolies and will bring about increasing competition, even if the formal regulatory regime
stays intact.
Vondracek et al (2000) in a journal on a statistical model for the estimation of natural gas
consumption presented a statistical approach to natural gas consumption estimation of individual
residential and small commercial customers. The approach is based on nonlinear regression
principles. Parameters are estimated using mainly two real data sets ordinary (approximately
annual) meter readings of almost all customers and additional (approximately monthly) meter
readings designed and operated within the frame of cooperation between the Institute of
Computer Science of the Czech Academy of Sciences (ICS) and the West Bohemian Gas
Distribution Company, a part of the RWE Group (WBG). The model was tested on various data
sets. It has broad applicability in many areas of gas industry.
Lorenson and Kvenvolden (2001) in a journal on the global occurrence of natural gas hydrate
noted that Natural gas hydrate occurs worldwide in oceanic sediment of continental and insular
slopes and rises of active and passive margins, in deep-water sediment of inland lakes and seas,
and in polar sediment on both continents and continental shelves. In aquatic sediment, where
water depths exceed about 300 m and bottom water temperatures approach 0C, gas hydrate is
found at the seafloor to sediment depths of about 1,100 m. In polar continental regions, gas
hydrate can be present in sediment at depths between about 150 and 2000 m. Thus, natural gas
4

hydrate is restricted to the shallow geosphere where its presence affects the physical and
chemical properties of near-surface sediment. An updated global inventory reports on natural gas
hydrate recovered from 19 places worldwide and includes 77 places where the presence of gas
hydrate has been inferred from geophysical, geochemical, and geological evidence. The potential
amount of methane in natural gas hydrate is enormous, with current estimates converging around
about 10 teratonnes (10
19
g) of methane carbon.
Smith et al (2001) in an article on natural gas production from hydrate decomposition by
depressurization studied natural gas production from the decomposition of methane hydrate in a
confined reservoir by a depressurizing well. The one-dimensional linearized model suggested by
Makogon was used in the analysis. For different well pressures and reservoir temperatures,
distributions of temperature and pressure in the porous layer of methane hydrate and in the gas
region are evaluated. The distance of the decomposition front from the well and the natural gas
output as functions of time are also computed. Time evolutions of the resulting temperature and
pressure profiles in the hydrate reservoir under various conditions are presented. Effects of
variations in reservoir porosity and zone permeability are also studied. It is shown that the gas
production rate is a sensitive function of well pressure, reservoir temperature and zone
permeability.
Siddiqi (2002) in a journal on natural gas reserves and total energy consumption noted that
Energy analysts have used the reserves/production ratios for oil and natural gas for decades as
indicators of the ability of countries to maintain or increase their production of those fuels. The
global community is now faced with the challenge of reducing carbon dioxide emissions from a
variety of sources, with the energy sector being the largest contributor to the anthropogenic
emissions of greenhouse gases. Natural gas has emerged as a highly desirable fuel, since it
produces lower emissions of carbon dioxide than coal or oil for equivalent amounts of energy
supplied. The ratio of a country's proven natural gas reserves to its total energy consumption is a
good indicator of its ability to improve its air quality situation or address greenhouse gas
reduction targets from domestic natural gas sources. This paper provides the ratio for several
countries at different stages of development, and discusses some of the implications. In countries
where exploration for natural gas has been limited, the estimated resources in place may
sometimes be a more useful indicator than proven reserves, and could be used instead.
Rothfarb et al (2003) in a journal on optimal design of offshore natural gas pipeline systems
noted natural gas reserves involves several phases, including production from reservoirs,
separation of byproducts, and transportation to markets. The gas, which may originate as far as
100 miles from land, must be transported through pipelines to onshore delivery points. This
paper develops techniques for solving the following problems: (1) selection of pipe diameters in
a specified pipeline network to minimize the sum of investment and operation costs; (2) selection
of minimum-cost network structures, given gas-field location and flow requirements; (3) optimal
expansion of existing pipeline networks to include newly discovered gas fields. The techniques
5

incorporate procedures for globally optimizing pipeline diameters for fixed tree structures and
heuristic procedures for generating low-cost structures.
Neumann et al (2004) in an article on steam reforming of natural gas pointed-out that the
development of heat resistant permeation membranes has opened up new possibilities for the
conversion of fossil energy resources. In steam reforming of natural gas, such membranes even
permit a direct production of hydrogen at high temperatures during the conversion of feed
hydrocarbons. Further gas processing, such as required for reformer gas in existing hydrogen
production processes, is not necessary. Due to continuous hydrogen discharge directly in the
reformer tube, the chemical equilibrium of the occurring reactions becomes displaced towards
the products, resulting in more favorable process conditions and, consequently, in improved by
36% utilization of the feed hydrocarbons. At the same time, the hydrogen yield increases by
44%. The heat required, which is provided by a high temperature reactor, is 17% in excess of
that in conventional plants. It can be expected that the simplified process design will produce
substantial cost advantages over the existing processes for the production of hydrogen.
Rudolf and Gabriel (2005) in their article on examining the market power in natural gas market,
developed a mixed complementarily equilibrium model for the European natural gas market.
This model has producers as Cournot players with conjectured supply functions relative to their
rivals. As such, these producers can withhold production to increase downstream prices for
greater profits. The other players are taken to be perfectly competitive and are combined with
extensive pipeline, seasonal, and other data reflecting the current state of the market. Four
market scenarios are run to analyze the extent of market power by these producers as well as the
importance of pipeline and storage capacity.
Ross et al (2005) in an article on the catalytic conversion of natural gas summarizes some
processes, direct and indirect, for the conversion of natural gas to useful products. It then
proceeds to give an outline of some work from the authors' laboratories on subjects such as
steam reforming, oxidative coupling and CO
2
reforming of methane, paying particular attention
to the development of novel catalysts for these processes.
Ubeda et al (2007) in an article on modeling and forecasting industrial end-use natural gas
consumption noted that Forecasting industrial end-use natural gas consumption is an important
prerequisite for efficient system operation and a basis for planning decisions. They presented a
novel prediction model that provides forecasting in a medium-term horizon (13 years) with a
very high resolution (days) based on a decomposition approach. The forecast is obtained by the
combination of three different components: one that captures the trend of the time series, a
seasonal component based on the Linear Hinges Model, and a transitory component to estimate
daily variations using explanatory variables. The flexibility of the model allows describing
demand patterns in a very wide range of historical profiles. Furthermore, the proposed method
combines a very simple representation of the forecasting model, which allows the expert to
6

integrate judgmental analysis and adjustment of the statistical forecast, with accuracy and high
computational efficiency. Realistic case studies are provided.
Egging et al (2008) in their article on a complementarily model for the natural gas market
present a detailed and comprehensive complementarily model for computing market equilibrium
values in the European natural gas system. Market players include producers and their marketing
arms which we call traders, pipeline and storage operators, marketers, LNG liquefiers,
regasifiers, tankers, and three end-use consumption sectors. The economic behavior of
producers, traders, pipeline and storage operators, liquefiers and regasifiers is modeled via
optimization problems whose KarushKuhnTucker (KKT) optimality conditions in
combination with market-clearing conditions form the complementarily system. The LNG
tankers, marketers and consumption sectors are modeled implicitly via appropriate cost
functions, aggregate demand curves, and ex post calculations, respectively. The model is run on
several case studies that highlight its capabilities, including a simulation of a disruption of
Russian supplies via Ukraine.
Groote (2009) Synthesis gas production from natural gas summarized that the feasibility of
producing synthesis gas by partial oxidation of natural gas on a Ni-catalyst in a fixed bed reactor
with reversed flow was investigated by means of simulation. A one dimensional reactor model of
the non-steady state heterogeneous type, accounting for internal diffusional limitations, was
applied. A double temperature peak is observed just after flow reversal, influencing the
selectives at the exit through enhanced steam reforming reactions. The second peak gradually
decreases in importance with time during the semi-cycle. The influence of several operating
conditions on the reactor performance was studied. Most of the coke deposited in a semi-cycle
can be removed after flow reversal. This observation opens new perspectives for the reversed
flow operation.









7

CHAPTER TWO
DATA COLLECTION
2.1 METHOD OF DATA COLLECTION

The data collected in this research is primary data It is a term for data collected from a source.
The data have not been subjected to processing or any other manipulation, and are also referred
to as raw data. Primary data can be input to a computer program or used in manual procedures
such as analyzing statistics from a survey.

2.2 DATA COLLECTED

2.2.1 PRODUCTION
This entry is the total natural gas produced in cubic meters (cu m). The discrepancy between the
amount of natural gas produced and/or imported and the amount consumed and/or exported is
due to the omission of stock changes and other complicating factors.
2.2.2 CONSUMPTION
This entry is the total natural gas consumed in cubic meters (cu m). The discrepancy between the
amount of natural gas produced and/or imported and the amount consumed and/or exported is
due to the omission of stock changes and other complicating factors.
2.2.3 EXPORTS
This entry is the total natural gas exported in cubic meters (cu m).
2.2.4 RESERVES
This entry is the stock of proved reserves of natural gas in cubic meters (cu m). Proved reserves
are those quantities of natural gas, which, by analysis of geological and engineering data, can be
estimated with a high degree of confidence to be commercially recoverable from a given date
forward, from known reservoirs and under current economic conditions.

2.3 HISTORY
Oil was discovered in Nigeria in 1956 at Oloibiri in the Niger Delta after half a century of
exploration. The discovery was made by Shell-BP, at the time the sole concessionaire.
Nigeria joined the ranks of oil producers in 1958 when its first oil field came on stream
8

producing 5,100 bpd. After 1960, exploration rights in onshore and offshore areas
adjoining the Niger Delta were extended to other foreign companies. In 1965 the EA field was
discovered by Shell in shallow water southeast of Warri. In 1970, the end of the Biafran war
coincided with the rise in the world oil price, and Nigeria was able to reap instant riches from its
oil production. Nigeria joined the Organization of Petroleum Exporting Countries (OPEC) in
1971 and established the Nigerian National Petroleum Company (NNPC) in 1977, a state owned
and controlled company which is a major player in both the upstream and downstream sectors.
Following the discovery of crude oil by Shell DArcy Petroleum, pioneer production began in
1958 from the companys oil field in Oloibiri in the Eastern Niger Delta. By the late sixties and
early seventies, Nigeria had attained a production level of over 2 million barrels of crude oil a
day. Although production figures dropped in the eighties due to economic slump, 2004 saw a
total rejuvenation of oil production to a record level of 2.5 million barrels per day. Current
development strategies are aimed at increasing production to 4million barrels per day by the
year 2010. Petroleum production and export play a dominant role in Nigeria's economy and
account for about 90% of her gross earnings. This dominant role has pushed agriculture, the
traditional mainstay of the economy, from the early fifties and sixties, to the background.

The discovery of oil opened up the Oil industry in 1961, bringing in Mobil, Agip, Safrap (now
Elf), Tenneco and Amoseas (Texaco and Chevron respectively) to join the exploration efforts
both in the onshore and areas of Nigeria. This development was enhanced by the extension of the
concessionary rights previously a monopoly of Shell, to the newcomers. The objective of the
government in doing this, was to he pace of exploration and production of Petroleum. Even now
more companies, both foreign and indigenous have won concessionary rights and are producing.
Actual oil production and export from the Oloibiri field in present day Bayelsa State commenced
in 1958 with an initial production rate of 5,100 barrels of crude oil per day. Subsequently, the
quantity doubled the following year and progressively as more players came onto the oil scene,
the production rose to 2.0 million barrels per day in 1972 and a peaking at 2.4 million barrels per
day in 1979. Nigeria thereafter, attained the status of a major oil producer, ranking 7th in the
world in 1972, and has since grown to become the sixth largest oil producing country in the
world.







9

CHAPTER THREE
3.0 DATA ANALYSIS

The analysis was run using SPSS V21.
Year Prod. Con. Expts Resv YEAR_ DATE_
Pred.
Prod.
Forcst.
Prod.
Pred.
Con.
Forcst.
Con.
Pred.
Expts.
Forecast
Expts.
Pred.
Resv.
Forcst
Resv.
1962 1780 110 1230 760800 1962 1962 -1176.5 131346.7
1963 2070 180 1540 668870 1963 1963 1659.49 146.52 1414.39 686695.8
1964 2830 270 1870 801820 1964 1964 1971.45 217.51 1810.76 700697
1965 3200 230 2000 654900 1965 1965 2741.37 316.69 2161.24 801401.3
1966 3540 200 2110 599900 1966 1966 3133.25 317.24 2363.13 754727.2
1967 3890 550 2430 789000 1967 1967 3496.93 288.06 2513.59 729201.3
1968 4400 700 1760 800800 1968 1968 3871.27 565.45 2796.38 854225.9
1969 4780 880 2310 832100 1969 1969 4401.78 792.7 2427.4 907901.2
1970 5030 900 2590 768900 1970 1970 4807.27 959.61 2737.79 956873.2
1971 5800 1050 2760 876800 1971 1971 5087.8 1032.58 3033.41 948358.8
1972 6540 1230 2980 1101000 1972 1972 5872.89 1153.11 3249.15 1028810
1973 6300 1110 3440 990980 1973 1973 6630.04 1347.35 3479.26 1199061
1974 6420 1450 3890 1100000 1974 1974 6440.2 1304.95 3879.14 1190379
1975 6510 1870 2570 888000 1975 1975 6599.35 1534.6 4315.11 1271461
1976 6620 1770 3480 990860 1976 1976 6730.49 1978.26 3535 1181479
1977 6790 2020 2990 1001000 1977 1977 6881.95 2003.17 3990.42 1240261
1978 7080 2440 4100 938530 1978 1978 7092.35 2155.56 3789.78 1277772
1979 7010 2230 3330 876880 1979 1979 7419.64 2575.61 4530.47 1267068
1980 7090 2580 4000 1111000 1980 1980 7400.58 2521.42 4208.28 1245809
1981 7670 2840 4320 1543000 1981 1981 7527.19 2718.45 4613.59 1398685
1982 7780 3030 5500 1550550 1982 1982 8137.04 3051.43 4960.95 1709566
1983 7890 3980 4320 1889000 1983 1983 8294.97 3252.02 5888.95 1803277
1984 8390 3300 4890 2002000 1984 1984 8453.93 4041.05 5330.41 2044763
1985 8200 3670 4330 1998800 1985 1985 8989.7 3784.06 5606.61 2185524
1986 8880 3450 5990 2997650 1986 1986 8862.06 3821.93 5312.1 2233850
1987 9030 3880 5440 2397000 1987 1987 9573.1 3833.77 6409.73 2859148
1988 9860 3900 5080 2932440 1988 1988 9775.2 4070.99 6330.27 2661831
1989 9760 3720 5420 1876990 1989 1989 10632.79 4246.47 6084.1 2954243
1990 10890 4020 5830 2867000 1990 1990 10596.92 4085.31 6280.24 2409195
1991 11100 4410 6020 3006000 1991 1991 11745.14 4285.5 6636.9 2893299
1992 11450 4550 6600 3760000 1992 1992 12010.08 4694.19 6881.95 3105153
10

1993 13800 5050 6980 4098760 1993 1993 12410.7 4899.36 7368.68 3621753
1994 11900 5320 5540 4204330 1994 1994 14734.84 5314.79 7778.21 3960291
1995 12220 4900 7010 3879000 1995 1995 12976.7 5671.89 6908.32 4118847
1996 13390 5800 7670 3987990 1996 1996 13351.75 5398.96 7728.65 3978963
1997 14430 4890 6790 4320600 1997 1997 14544.42 5976.5 8416.52 4029973
1998 12890 6570 6540 4008700 1998 1998 15613.69 5571.54 8004.8 4258193
1999 15500 6120 7340 3787890 1999 1999 14207.11 6550.43 7750.58 4142283
2000 14860 6790 7500 3990000 2000 2000 16784.9 6774.74 8263.11 4001920
2001 15680 7850 7830 3986000 2001 2001 16245.31 7032.44 8526.88 4108406
2002 17680 7560 9580 4007000 2002 2002 17107.55 8123.38 8845.3 4148403
2003 19500 7410 10950 4007000 2003 2003 19103.24 8143.97 10160.53 4188094
2004 22100 9210 12590 4007000 2004 2004 20928 7896.26 11465.28 4215179
2005 26420 10770 15230 4984000 2005 2005 23502.53 9294.13 12954.26 4239377
2006 30250 11650 18100 5100000 2006 2006 27728.37 11024.81 15182.16 4846105
2007 34100 12900 21200 5198700 2007 2007 31486.92 12028.22 17756.48 5072319
2008 32820 12280 20550 5210000 2008 2008 35266.93 13176.48 20577.8 5201044
2009 35210 12280 28800 5216800 2009 2009 34132.06 13018.55 20865.84 5255269
2010 34540 12010 21760 5246000 2010 2010 36514.74 12773.58 26672.84 5289742
2011 35450 12430 24500 5347890 2011 2011 36298.55 12990.8 28430.66 5324361
2012 2012 36205.13 13046.37 23523.04 5412948
2013 2013 36940.48 13573.99 23762.72 5496926
2014 2014 37617.8 14056.24 24012.51 5584284
2015 2015 38241.66 14573.83 24271.88 5674433
2016 2016 39640.98 15167.76 24588 5798976









11

3.1 PRODUCTION

Sequence Plot
Model Description
Model Name MOD_1
Series or Sequence 1 Production
Transformation None
Non-Seasonal Differencing 0
Seasonal Differencing 0
Length of Seasonal Period No periodicity
Horizontal Axis Labels Date_
Intervention Onsets None
Reference Lines None
Area Below the Curve Not filled
Applying the model specifications from MOD_1

Case Processing Summary

Production
Series or Sequence Length 50
Number of Missing Values in
the Plot
User-Missing 0
System-Missing 0



The production variables shows an upward trend. Theres a-bit of seasonality in the general data.
12

ACF

Model Description
Model Name MOD_2
Series Name 1 Production
Transformation None
Non-Seasonal Differencing 1
Seasonal Differencing 0
Length of Seasonal Period No periodicity
Maximum Number of Lags 16
Process Assumed for Calculating the Standard Errors of the
Autocorrelations
Independence(white noise)
a

Display and Plot All lags
Applying the model specifications from MOD_2
a. Not applicable for calculating the standard errors of the partial autocorrelations.





























Case Processing Summary

Production
Series Length 50
Number of Missing Values
User-Missing 0
System-Missing 0
Number of Valid Values 50
Number of Values Lost Due to Differencing 1
Number of Computable First Lags After Differencing 47
13

Production
Autocorrelations
Series: Production
Lag Autocorrelation Std. Error
a
Box-Ljung Statistic
Value df Sig.
b

1 .152 .140 1.180 1 .277
2 .395 .138 9.336 2 .009
3 .216 .137 11.829 3 .008
4 .133 .135 12.794 4 .012
5 -.075 .134 13.105 5 .022
6 .095 .132 13.623 6 .034
7 -.031 .131 13.681 7 .057
8 -.018 .129 13.701 8 .090
9 -.046 .127 13.831 9 .128
10 .102 .126 14.492 10 .152
11 -.120 .124 15.420 11 .164
12 .020 .122 15.447 12 .218
13 -.010 .121 15.453 13 .280
14 .091 .119 16.036 14 .311
15 -.070 .117 16.395 15 .356
16 .079 .115 16.864 16 .394
a. The underlying process assumed is independence (white noise).
b. Based on the asymptotic chi-square approximation.


14

Partial Autocorrelations
Series: Production
Lag Partial
Autocorrelation
Std. Error
1 .152 .144
2 .381 .144
3 .145 .144
4 -.052 .144
5 -.263 .144
6 .057 .144
7 .099 .144
8 .001 .144
9 -.093 .144
10 .108 .144
11 -.072 .144
12 -.046 .144
13 .021 .144
14 .174 .144
15 -.054 .144
16 -.093 .144





15

Sequence Plot
Model Description
Model Name MOD_3
Series or Sequence
1 Production
2
Predicted value from Production-
Model_1
Transformation None
Non-Seasonal Differencing 0
Seasonal Differencing 0
Length of Seasonal Period No periodicity
Horizontal Axis Labels Date_
Intervention Onsets None
For Each Observation Values not joined
Applying the model specifications from MOD_3

Case Processing Summary

Production Predicted value
from Production-
Model_1
Series or Sequence Length 54 54
Number of Missing Values in
the Plot
User-Missing 0 0
System-Missing 5 6




16

Apply Time Series Models

Model Description

Model Type
Model ID Production Model_1 ARIMA(1,1,1)


Model Fit
Fit Statistic Mean SE Minimum Maximum Percentile
5 10 25 50 75 90 95
Stationary
R-squared
.053 . .053 .053 .053 .053 .053 .053 .053 .053 .053
R-squared .981 . .981 .981 .981 .981 .981 .981 .981 .981 .981
RMSE 1248.600 . 1248.600 1248.600 1248.600 1248.600 1248.600 1248.600 1248.600 1248.600 1248.600
MAPE 6.093 . 6.093 6.093 6.093 6.093 6.093 6.093 6.093 6.093 6.093
MaxAPE 22.636 . 22.636 22.636 22.636 22.636 22.636 22.636 22.636 22.636 22.636
MAE 814.207 . 814.207 814.207 814.207 814.207 814.207 814.207 814.207 814.207 814.207
MaxAE 3662.981 . 3662.981 3662.981 3662.981 3662.981 3662.981 3662.981 3662.981 3662.981 3662.981
Normalized
BIC
14.421 . 14.421 14.421 14.421 14.421 14.421 14.421 14.421 14.421 14.421


The largest forecasted error is 3662.98 and expressed in percentage as 22.636%

The developed model varies from the original with 6.093%

The overall fit of a model that attempts to account for model complexity is 14.421% < 19% on
the average.

An R-square value of .961 shows that the developed model is better than the original model.

Model Statistics
Model Number of
Predictors
Model Fit statistics Ljung-Box Q(18) Number of
Outliers
Stationary R-
squared
Statistics DF Sig.
Production-Model_1 0 .053 15.351 16 .499 0


-value > 0.05 shows that theres no correlation between the first 18-lags.





17


Forecast
Model 2012 2013 2014 2015 2016
Production-Model_1
Forecast 36205.13 36940.48 37617.80 38241.66 39640.98
UCL 40190.37 42330.82 44404.20 46427.58 48329.5
LCL 32219.88 31550.14 30831.40 30055.74 29998.8
For each model, forecasts start after the last historical period that was used in estimation of the models applied,
and end at the last period for which non-missing values of all the predictors are available or at the end date of the
requested forecast period, whichever is earlier.







18














19

3.2 CONSUMPTION
Sequence Plot
Model Description
Model Name MOD_4
Series or Sequence 1 Consumption
Transformation None
Non-Seasonal Differencing 0
Seasonal Differencing 0
Length of Seasonal Period No periodicity
Horizontal Axis Labels Date_
Intervention Onsets None
Reference Lines None
Area Below the Curve Not filled
Applying the model specifications from MOD_4

Case Processing Summary

Consumption
Series or Sequence Length 54
Number of Missing Values in
the Plot
User-Missing 0
System-Missing 5






20


ACF
Model Description
Model Name MOD_5
Series Name 1 Consumption
Transformation None
Non-Seasonal Differencing 1
Seasonal Differencing 0
Length of Seasonal Period No periodicity
Maximum Number of Lags 16
Process Assumed for Calculating the Standard Errors of the
Autocorrelations
Independence(white noise)
a

Display and Plot All lags
Applying the model specifications from MOD_5
a. Not applicable for calculating the standard errors of the partial autocorrelations.

Case Processing Summary

Consumption
Series Length 54
Number of Missing Values
User-Missing 0
System-Missing 5
a

Number of Valid Values 50
Number of Values Lost Due to Differencing 1
Number of Computable First Lags After Differencing 47
a. Some of the missing values are imbedded within the series.




























21

Consumption
Autocorrelations
Series: Consumption
Lag Autocorrelation Std. Error
a
Box-Ljung Statistic
Value df Sig.
b

1 -.171 .140 1.486 1 .223
2 .197 .138 3.501 2 .174
3 -.002 .137 3.502 3 .321
4 -.151 .135 4.748 4 .314
5 .015 .134 4.761 5 .446
6 .089 .132 5.209 6 .517
7 -.001 .131 5.209 7 .634
8 .041 .129 5.312 8 .724
9 .012 .127 5.320 9 .806
10 -.204 .126 7.941 10 .635
11 .216 .124 10.975 11 .445
12 -.270 .122 15.826 12 .199
13 .254 .121 20.246 13 .089
14 -.117 .119 21.217 14 .096
15 .056 .117 21.448 15 .123
16 -.031 .115 21.521 16 .159
a. The underlying process assumed is independence (white noise).
b. Based on the asymptotic chi-square approximation.





22

Partial Autocorrelations
Series: Consumption
Lag Partial
Autocorrelation
Std. Error
1 -.171 .144
2 .172 .144
3 .059 .144
4 -.188 .144
5 -.048 .144
6 .170 .144
7 .052 .144
8 -.049 .144
9 -.005 .144
10 -.175 .144
11 .197 .144
12 -.172 .144
13 .144 .144
14 -.084 .144
15 .053 .144
16 -.043 .144







23

Sequence Plot
Model Description
Model Name MOD_6
Series or Sequence
1 Consumption
2
Predicted value from
Consumption-Model_1
Transformation None
Non-Seasonal Differencing 0
Seasonal Differencing 0
Length of Seasonal Period No periodicity
Horizontal Axis Labels Date_
Intervention Onsets None
For Each Observation Values not joined
Applying the model specifications from MOD_6

Case Processing Summary

Consumption Predicted value
from
Consumption-
Model_1
Series or Sequence Length 54 54
Number of Missing Values in
the Plot
User-Missing 0 0
System-Missing 5 1




24

Apply Time Series Models

Model Description

Model Type
Model ID Consumption Model_1 ARIMA(1,1,1)




Model Summary

Model Fit
Fit Statistic Mean SE Minimum Maximum Percentile
5 10 25 50 75 90 95
Stationary R-
squared
.107 . .107 .107 .107 .107 .107 .107 .107 .107 .107
R-squared .978 . .978 .978 .978 .978 .978 .978 .978 .978 .978
RMSE 560.342 . 560.342 560.342 560.342 560.342 560.342 560.342 560.342 560.342 560.342
MAPE 11.707 . 11.707 11.707 11.707 11.707 11.707 11.707 11.707 11.707 11.707
MaxAPE 58.620 . 58.620 58.620 58.620 58.620 58.620 58.620 58.620 58.620 58.620
MAE 383.215 . 383.215 383.215 383.215 383.215 383.215 383.215 383.215 383.215 383.215
MaxAE 1475.873 . 1475.873 1475.873 1475.873 1475.873 1475.873 1475.873 1475.873 1475.873 1475.873
Normalized
BIC
12.980 . 12.980 12.980 12.980 12.980 12.980 12.980 12.980 12.980 12.980


The largest forecasted error is 1475.873 and expressed in percentage as 58.62%

The developed model varies from the original with 11.707%

The overall fit of a model that attempts to account for model complexity is 12.98% < 19% on the
average.

An R-square value of .978 shows that the developed model is better than the original model.



Model Statistics
Model Number of
Predictors
Model Fit statistics Ljung-Box Q(18) Number of
Outliers
Stationary R-
squared
Statistics DF Sig.
Consumption-Model_1 1 .107 15.828 16 .465 0

-value > 0.05 shows that theres no correlation between the first 18-lags.
25

Forecast
Model 2012 2013 2014 2015 2016
Consumption-Model_1
Forecast 13046.37 13573.99 14056.24 14573.83 15167.76
UCL 14474.44 15315.72 16034.41 16775.45 17864.9
LCL 11618.31 11832.25 12078.07 12372.21 13121.88
For each model, forecasts start after the last historical period that was used in estimation of the models applied, and
end at the last period for which non-missing values of all the predictors are available or at the end date of the
requested forecast period, whichever is earlier.







26
















27

3.3 EXPORT
Sequence Plot
Model Description
Model Name MOD_7
Series or Sequence 1 Exports
Transformation None
Non-Seasonal Differencing 0
Seasonal Differencing 0
Length of Seasonal Period No periodicity
Horizontal Axis Labels Date_
Intervention Onsets None
Reference Lines None
Area Below the Curve Not filled
Applying the model specifications from MOD_7

Case Processing Summary

Exports
Series or Sequence Length 54
Number of Missing Values in
the Plot
User-Missing 0
System-Missing 5




28

ACF
Model Description
Model Name MOD_8
Series Name 1 Consumption
Transformation None
Non-Seasonal Differencing 1
Seasonal Differencing 0
Length of Seasonal Period No periodicity
Maximum Number of Lags 16
Process Assumed for Calculating the Standard Errors of the
Autocorrelations
Independence(white noise)
a

Display and Plot All lags
Applying the model specifications from MOD_8
a. Not applicable for calculating the standard errors of the partial autocorrelations.

Case Processing Summary

Consumption
Series Length 54
Number of Missing Values
User-Missing 0
System-Missing 5
a

Number of Valid Values 50
Number of Values Lost Due to Differencing 1
Number of Computable First Lags After Differencing 47
a. Some of the missing values are imbedded within the series.



























29

Consumption
Autocorrelations
Series: Consumption
Lag Autocorrelation Std. Error
a
Box-Ljung Statistic
Value df Sig.
b

1 -.171 .140 1.486 1 .223
2 .197 .138 3.501 2 .174
3 -.002 .137 3.502 3 .321
4 -.151 .135 4.748 4 .314
5 .015 .134 4.761 5 .446
6 .089 .132 5.209 6 .517
7 -.001 .131 5.209 7 .634
8 .041 .129 5.312 8 .724
9 .012 .127 5.320 9 .806
10 -.204 .126 7.941 10 .635
11 .216 .124 10.975 11 .445
12 -.270 .122 15.826 12 .199
13 .254 .121 20.246 13 .089
14 -.117 .119 21.217 14 .096
15 .056 .117 21.448 15 .123
16 -.031 .115 21.521 16 .159
a. The underlying process assumed is independence (white noise).
b. Based on the asymptotic chi-square approximation.




30

Partial Autocorrelations
Series: Consumption
Lag Partial
Autocorrelation
Std. Error
1 -.171 .144
2 .172 .144
3 .059 .144
4 -.188 .144
5 -.048 .144
6 .170 .144
7 .052 .144
8 -.049 .144
9 -.005 .144
10 -.175 .144
11 .197 .144
12 -.172 .144
13 .144 .144
14 -.084 .144
15 .053 .144
16 -.043 .144




31

Sequence Plot
Model Description
Model Name MOD_9
Series or Sequence
1 Exports
2
Predicted value from Exports-
Model_1
Transformation None
Non-Seasonal Differencing 0
Seasonal Differencing 0
Length of Seasonal Period No periodicity
Horizontal Axis Labels Date_
Intervention Onsets None
For Each Observation Values not joined
Applying the model specifications from MOD_9


Case Processing Summary

Exports Predicted value
from Exports-
Model_1
Series or Sequence Length 54 54
Number of Missing Values in
the Plot
User-Missing 0 0
System-Missing 5 0


32


Apply Time Series Models

Model Description

Model Type
Model ID Exports Model_1 ARIMA(1,0,1)


Model Fit
Fit Statistic Mean SE Minimum Maximum Percentile
5 10 25 50 75 90 95
Stationary
R-squared
.915 . .915 .915 .915 .915 .915 .915 .915 .915 .915
R-squared .915 . .915 .915 .915 .915 .915 .915 .915 .915 .915
RMSE 1812.264 . 1812.264 1812.264 1812.264 1812.264 1812.264 1812.264 1812.264 1812.264 1812.264
MAPE 18.215 . 18.215 18.215 18.215 18.215 18.215 18.215 18.215 18.215 18.215
MaxAPE 195.651 . 195.651 195.651 195.651 195.651 195.651 195.651 195.651 195.651 195.651
MAE 1029.441 . 1029.441 1029.441 1029.441 1029.441 1029.441 1029.441 1029.441 1029.441 1029.441
MaxAE 7934.156 . 7934.156 7934.156 7934.156 7934.156 7934.156 7934.156 7934.156 7934.156 7934.156
Normalized
BIC
15.322 . 15.322 15.322 15.322 15.322 15.322 15.322 15.322 15.322 15.322

The largest forecasted error is 7934.156 and expressed in percentage as 195.561%

The developed model varies from the original with 18.215%

The overall fit of a model that attempts to account for model complexity is 15.322% < 19% on
the average.

An R-square value of .915 shows that the developed model is better than the original model.


Model Statistics
Model Number of
Predictors
Model Fit statistics Ljung-Box Q(18) Number of
Outliers
Stationary R-
squared
Statistics DF Sig.
Exports-Model_1 1 .915 11.441 16 .782 0

-value > 0.05 shows that theres no correlation between the first 18-lags.


33


Forecast
Model 2012 2013 2014 2015 2016
Exports-Model_1
Forecast 23523.04 23762.72 24012.51 24271.88 24588
UCL 27890.49 28722.53 29447.59 30099.17 30234.11
LCL 19155.58 18802.90 18577.44 18444.58 18123.58
For each model, forecasts start after the last historical period that was used in estimation of the models
applied, and end at the last period for which non-missing values of all the predictors are available or at the end
date of the requested forecast period, whichever is earlier.





34




















35

3.4 RESERVE
Sequence Plot
Model Description
Model Name MOD_10
Series or Sequence 1 Reserve
Transformation None
Non-Seasonal Differencing 0
Seasonal Differencing 0
Length of Seasonal Period No periodicity
Horizontal Axis Labels Date_
Intervention Onsets None
Reference Lines None
Area Below the Curve Not filled
Applying the model specifications from MOD_10

Case Processing Summary

Reserve
Series or Sequence Length 54
Number of Missing Values in
the Plot
User-Missing 0
System-Missing 5



36

ACF

Model Description
Model Name MOD_11
Series Name 1 Reserve
Transformation None
Non-Seasonal Differencing 1
Seasonal Differencing 0
Length of Seasonal Period No periodicity
Maximum Number of Lags 16
Process Assumed for Calculating the Standard Errors of the
Autocorrelations
Independence(white noise)
a

Display and Plot All lags
Applying the model specifications from MOD_11
a. Not applicable for calculating the standard errors of the partial autocorrelations.

Case Processing Summary

Reserve
Series Length 54
Number of Missing Values
User-Missing 0
System-Missing 5
a

Number of Valid Values 49
Number of Values Lost Due to Differencing 1
Number of Computable First Lags After Differencing 47
a. Some of the missing values are imbedded within the series.


























37

Reserve
Autocorrelations
Series: Reserve
Lag Autocorrelation Std. Error
a
Box-Ljung Statistic
Value df Sig.
b

1 -.395 .140 7.972 1 .005
2 .324 .138 13.434 2 .001
3 -.349 .137 19.919 3 .000
4 .118 .135 20.680 4 .000
5 -.028 .134 20.723 5 .001
6 -.023 .132 20.754 6 .002
7 -.014 .131 20.765 7 .004
8 -.093 .129 21.287 8 .006
9 .046 .127 21.419 9 .011
10 .029 .126 21.472 10 .018
11 -.004 .124 21.473 11 .029
12 .070 .122 21.795 12 .040
13 -.034 .121 21.876 13 .057
14 .118 .119 22.864 14 .063
15 .002 .117 22.864 15 .087
16 -.145 .115 24.451 16 .080
a. The underlying process assumed is independence (white noise).
b. Based on the asymptotic chi-square approximation.



38

Partial Autocorrelations
Series: Reserve
Lag Partial
Autocorrelation
Std. Error
1 -.395 .144
2 .198 .144
3 -.207 .144
4 -.132 .144
5 .114 .144
6 -.088 .144
7 -.099 .144
8 -.072 .144
9 -.033 .144
10 .055 .144
11 -.035 .144
12 .057 .144
13 .056 .144
14 .090 .144
15 .118 .144
16 -.207 .144




39


Time Series Modeler

Model Description

Model Type
Model ID Reserve Model_1 ARIMA(1,0,1)


Model Summary

Model Fit
Fit Statistic Mean SE Minimum Maximum Percentile
5 10 25 50 75 90 L95
Stationary
R-squared
.959 . .959 .959 .959 .959 .959 .959 .959 .959 .959
R-squared .959 . .959 .959 .959 .959 .959 .959 .959 .959 .959
RMSE 340789.867 . 340789.867 340789.867 340789.867 340789.867 340789.867 340789.867 340789.867 340789.867 340789.867
MAPE 14.041 . 14.041 14.041 14.041 14.041 14.041 14.041 14.041 14.041 14.041
MaxAPE 82.736 . 82.736 82.736 82.736 82.736 82.736 82.736 82.736 82.736 82.736
MAE 238973.986 . 238973.986 238973.986 238973.986 238973.986 238973.986 238973.986 238973.986 238973.986 238973.986
MaxAE 1077252.923 . 1077252.923 1077252.923 1077252.923 1077252.923 1077252.923 1077252.923 1077252.923 1077252.923 1077252.923
Normalized
BIC
25.796 . 25.796 25.796 25.796 25.796 25.796 25.796 25.796 25.796 25.796

The largest forecasted error is 1077252.923 and expressed in percentage as 82.736%

The developed model varies from the original with 14.041%

The overall fit of a model that attempts to account for model complexity is 25.796% > 19% on
the average.

An R-square value of .959 shows that the developed model is better than the original model.


Model Statistics
Model Number of
Predictors
Model Fit statistics Ljung-Box Q(18) Number of
Outliers
Stationary R-
squared
Statistics DF Sig.
Reserve-Model_1 1 .959 17.667 16 .344 0

-value > 0.05 shows that theres no correlation between the first 18-lags.

40



Forecast
Model 2012 2013 2014 2015 2016
Reserve-Model_1
Forecast 5.41E+006 5.50E+006 5.58E+006 5.67E+006 5.79E+006
UCL 6.20E+006 6.35E+006 6.48E+006 6.60E+006 6.62E+006
LCL 4.63E+006 4.65E+006 4.69E+006 4.75E+006 4.80E+006
For each model, forecasts start after the last historical period that was used in estimation of the models applied, and end
at the last period for which non-missing values of all the predictors are available or at the end date of the requested
forecast period, whichever is earlier.






41





















42

CHAPTER FOUR
4.0 SUMMARY, CONCLUSION AND RECCOMMENDATION

Nigeria depends majorly on her activities in the oil and gas industry. She produces,
consumes, exports and reserves her oil. In studying these variables, a raw data
showing Nigerias level of production, consumption, quantity exported and
reserved from 1962 2011.
The analysis was run using SPSS v21.0 with the aim modeling the oil variables
over the years and making forecast to the future (2012-2016).
CONCLUSION
Analysis showed that although not steady, there was a continuous increase in the level of
production from 1962 2011. The model ARIMA(1,1,1) was developed for forecasting
subsequent years. The forecast showed there will be a steady rise in the production of oil in
Nigeria.
There was a continuous increase in the level of consumption from 1962 2011. The model
ARIMA(1,1,1) was developed for forecasting subsequent years. The forecast showed there will
be a steady rise in the production of oil in Nigeria.
There was a continuous increase in the level of export from 1962 2011. The model
ARIMA(1,0,1) was developed for forecasting subsequent years. The forecast showed there will
be a steady rise in the production of oil in Nigeria.
There was a continuous increase in the level of reserve from 1962 2011. The model
ARIMA(1,0,1) was developed for forecasting subsequent years. The forecast showed there will
be a steady rise in the production of oil in Nigeria.
RECOMMENDATION
With the constant steady rise shown by all the variables, the oil and gas Ministry should continue
strictly with her activities and if the need be for change, measures should be effectively verified.