NIGERIA (1988-2011)
ABSTRACT
This project is based on time series analysis of yearly sales of petroleum at Balewa fuel
station Nigeria limited, Oshogbo, Osun State from 1988-2011. Various trend models of
time series analysis were discussed with a view of showing understanding to the
appropriate method to be used for forecasts. The time plot of yearly sales of petroleum
shows that, there is low sales of petroleum within year 1994 and higher sales of
petroleum occur during year 2011. Moreover, evidence from trend analysis shows that
the quadratic model trend gives the best model out of all the models considered. This is
due to the model which has the least Mean Absolute Deviation. The study adopted
quadratic trend model, which was concluded that quadratic model was the best model
that best fit the data and was used to forecast yearly sales of petroleum from year 1998
to 2023.
1
TABLE OF CONTENTS
TITLE
PAGE
Title Page i
Certification ii
Dedication iii
Acknowledgement iv
Abstract v
Table of Contents
vi-vii
List of Tables
ix
List of Figures
x
2
CHAPTER ONE
1.0 Introduction
1
Conclusion of chapter 6
2.0 Introduction 7
3
2.5.1 Forecasting Accuracy
11
2.8.2 Correlogram 14
CHAPTER THREE
3.0 Introduction 17
3.2 Forecast
21
4.0 Introduction 23
4.2 Recommendations 23
References 24
Appendix 25
4
LIST OF TABLES
Table 3: Appendix 25
LIST OF FIGURES
5
CHAPTER ONE
INTRODUCTION
The search of oil in Nigeria started as early as 1937, but the discovery was not until
1956. The sole of petroleum products began in December 1957, managed by a
consortium of Royal Dutch Shell and British Petroleum BP Now known as Shell
Petroleum Development Company SPDC. G.A, Aga (1993) stated that Nigeria was the
second oil producing nation in Africa after Libya and sixth in the world.
In May 1971, the Nigeria National Oil Company was established under the
company and Allied matter Act of 1958 as applicable then. NNOC was the government
Agency Mandated by law to engage in all phases of oil production and sales, NNOC
was later in 1977 amalgamated into a full flex ministry of petroleum to form the Nigeria
National Petroleum Corporation (NNPC), which is in partnership with several oil
company from different countries operating in Nigeria. Before October 1965, Nigeria
Crude Oil was refined overseas and all the processed oil needs were imported. The first
refinery plant came into operation in 1965 located at Alesa Eleme near Port-Harcourt.
Later Warri and Kaduna Petro-chemical refineries were established in 1978 and 1980
respectively. Similarly, Pipeline and Products Marketing Company Ltd (PPMC) Enugu
Depot was commissioned in 25 August 1975 by the then military Governor of the old
Anambra State; Colonel D.S. Abubakar. The last was the second refinery in Port-
Harcourt. It is however worthy to note that NNPC has several subsidiary company e.g.
Pipeline and Product Marketing Company (PPMC).
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 producing 5,100 bpd. After 1960, exploration rights in onshore and
6
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 [Blair1976, pp. 98-120]. Following the discovery of crude oil by Shell D’Arcy
Petroleum, pioneer production began in 1958 from the company’s 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. While the discovery of
oil in the eastern and mid-western regions of the Niger Delta pleased hopeful Nigerians,
giving them an early indication soon after independent economic development was
within reach, at the same time it signaled a danger of grave consequence: oil revenues
fueled already existing ethnic and political tension and actually "burned" the country.
This tension reached its peak with the civil war that lasted from 1967 to 1970. As the
war commenced, the literature reflected the hostility, the impact, and fate of the oil
industry. Nigeria survived the war, and was able to recover mainly of the huge revenues
from oil in the 1970s. For some three years an oil boom followed, and the country was
awash with money.
Indeed, there was money for virtually all the items in its developmental plan.
The literature of the postwar years shifted to the analysis of the world oil boom and
bust, collectively known as the "oil shock." Starting in 1973 the world experienced an
oil shock that rippled through Nigeria until the mid - 1980s. This oil shock was initially
7
positive for the country, but with mismanagement and military rule, it became all
economic disaster. The larger middle class produced by the oil boom of the 1970s
gradually became disenchanted in the 1980s, and rebellious in the 1990s. The enormous
impact of the oil shock could not escape scholarly attention. For almost twenty years
(1970s 1990s), the virtual obsession was to analyze the consequences of oil on Nigeria,
using different models and theories. A set of radical-oriented writers was concerned
with the nationalization that took place during the oil shock as well as the linkages
between oil and an activist foreign policy.
Regarding the latter, the emphasis was on OPEC, Nigeria's strategic alliance formation
within Africa, the vigorous efforts to establish the Economic Community of West
African States (ECOWAS), and the country's attempts to use oil as a political weapon,
especially in the liberation of South Africa from apartheid. If many had hoped that oil
would turn Nigeria into an industrial power and a prosperous country based on a large
middle class, they were to be disappointed when a formally rich country became a
debtor nation by the 1980s. The suddenness of the economic difficulties of the 1980s
"bust years" had an adverse effect on class relations and the oil workers who understood
the dynamics of the industry. As if to capture the labor crisis, writings on oil workers
during this period covered many interrelated issues, notably working conditions, strikes,
and state labor relations. To be sure, labor issues were not new in the 1980s, since the
left-oriented scholars had made a point of exposing labor relations in the colonial era.
What was new after 1980 was the focus on oil workers, unions, and class conflict
[OPEC annual report 1983].
1. The first Port Harcourt Refinery was commissioned in 1965 with an installed
capacity of 35,000 bpd and later expanded to 60,000 bpd.
2. The Warri Refinery was commissioned in 1978 with an installed refining capacity
100,000 bpd, and upgraded to 125,000 bpd in 1986.
3. The Kaduna Refinery was commissioned in 1980 with an installed refining capacity
of 100,000 bpd, and upgraded to 110,000 bpd in 1986.
4. The second Port Harcourt Refinery was commissioned in1989 with 150,000 bpd
processing capacity, and designed to fulfil the dual role of supplying the domestic
market and exporting its surplus. The combined capacities of these refineries exceed the
9
domestic consumption of refined products, chief of which is premium motor spirit
(gasoline), whose demand is estimated at 33 million litres daily. The refineries are
however, operating far below their installed capacities, as they were more or less
abandoned during the military era, skipping the routine and mandatory turnaround
maintenance that made products shortages that gave strength to the argument for
deregulation of the downstream oil subsector in Nigeria. The monetization of oil
revenue has been a major factor in liquidity management in Nigeria. Measuring
liquidity as the narrow and broad money definitions by the CBN, the early 1990s saw
increases that were dampened by 1995 up until the civilian administration came on
board in 1999. The new Government maintained disciplined fiscal operations for about
one year and thereafter, the floodgates were opened. Since then, the CBN has been
battling to keep liquidity in check, in order to ensure that it does not create adverse
effects on the three key macroeconomic prices (i.e., interest rate, exchange rate and
inflation rate).
The greatest challenge is when Nigeria generates more revenue from crude oil sales
than it budgeted, like now. Such excesses have always been monetized, creating market
distortions and inflationary pressure [Biodun Adedipe 2004]. The same argument goes
for deficit fiscal operations in comparison to the GDP. The pattern of this ratio indicates
the optimism that accompanies increase in oil revenue and makes Government to
engage in frivolous spending or unnecessary projects. Deficit spending invariably
makes Government resort to borrowing from the Central Bank through the instrument
of Ways and Means Advances, which later convert into shortterm debt instruments that
are quite expensive to service at market rates. At this point, there is sufficient ground to
examine how economic policy formulation has been impacted or induced by petroleum
oil in Nigeria. As much as possible, major economic policies since Nigeria gained
political independence would be examined vis-а-vis the state of the oil sector. This
should provide adequate basis for making a few specific recommendations on how to
reduce the dependency. Importation inevitable, Importation not withstanding, there
have been persistent product.
1.3 NIGERIAN NATIONAL PETROLEUM CORPORATION (ITS ROLE IN
SALE OF PETROLEUM PRODUCTS)
10
The NNPC‟s role in Oil Industry is so much that it cannot handled it alone. This is the
reason for the establishment of subsidiary company like pipeline and Products
Marketing Company Ltd (PPMC). The Nigerian National Petroleum Corporation
manages the affairs of the oil industry in Nigeria, while the PPMC under the
corporation is in charge of sales of petroleum products. Government policy on oil
matter such sales is been conveyed by the Petroleum Products Price Regulatory Agency
(PPPRA) currently headed by Alhaji Gbalamosi. NNPC therefore, works in conjunction
with PPPRA to implement government policy such as prices of petroleum products.
Nigerian National Petroleum Corporation carries out its function as such in both local
and international.
This chapter discussed the yearly sales of petroleum in Oshogbo, discovery of crude oil
in Nigeria and the aim and objective of the study and the description of the data.
11
CHAPTER TWO
LITERATURE REVIEW
2.0 INTRODUCTION
This chapter is concerned with the literature review of past work on time series data and
the sample techniques used in analysis the data in chapter three.
A special feature of times series analysis is the fact that successive observations
usually take into account the time order of the observations. When successive
observations are dependent, future values may be predicted from past observations .If
the series can be predicted exactly, it is said to be deterministic. Most times series are
however stochastic in that future values are only partly determined by past values
having a probability distribution which is conditioned by knowledge of past values. A
time series can be either;
Time series can be classified into four different types such as:
1. Economic time series: This involves the study of changes in the economy of a
given community, nation or company e.g monthly data for unemployment.
2. Marketing time series: This concerned with the behaviour of finished products
in the market in terms of quality and sales over a period of time.
12
3. Physical time series: This is a time series arising from physical series such as
marine, geophysics, metrology,etc. This involves the recording of rainfall,
pressure, temperature, etc in time.
4. Demography time series: This involves the study of human population in a
given geographical area over a period of time.
1. Secular trend(Tt): This is a long term movement in the mean, it refer to the
general direction in which the graph of a time series appears to be going over a
long time interval of time,it is indicted by a trend line.The secular trend is
generally either upward or downward. General trend is the result of such force
which are more or less constant for a long time or which change gradually or
slowly. The determination of such trend lines and curves is by the least squares
method.
2. Cyclical variation (Ct): This is the long term oscillation or swing about a trend
line or curve. It may or may not be periodic due to various combinations of
factors influencing the economy.
4. Irregular variation (It): This is the remaining component after estimating the
trend seasonal and cyclical variations. These refer to the sporadic motions in the
time series due to change events such as floods strikes and elections. Although it
is ordinarily assumed that such events produce variations lasting only a short
time it is conceivable that they may be so intense as to result it new cyclical
variations from a set of data.
13
2.1.3 TIME SERIES TREND
Yt = βο + ( β1*t ) + et
In this model, β1 represents the average change from period to the next.
The quadratic model which can explain simple curvature in the data is:
Yt = βο + β1*t ( β2*t2 ) + et
The exponential growth trend model explains exponential growth or decay. For
example, a savings account might exhibit exponential growth. The model is:
Yt = βο + ( β1t ) + et
The S-curve trend model fits the pearl-reed logistic trend model, this explains the case
where the series follows an S-cure. The model is:
Yt = (10a) / ( βο + β1 β2t )
a. DISCRIPTION: When presented with time series, the first step in the analysis is
usually to plot the data to obtain simple descriptive measures of the main
properties of the series.
b. EXPLAINATION: When the observations are taken on two or more variables, it
may be possible to use the variation in one time series to explain the variation in
another series.
c. PREDICTION: Given an observed time series, one may want to predict the future
values of the series. This is important task in sales forecasting and in the analysis
of economic and industrial time series.
d. CONTROL: When time series is generated, which measures the quality of a
manufacturing process. The aim of the analysis may be to control the process. The
observations are plotted on control chart and the control lines take sanctions as a
result of studying the chart.
14
2.1.5 IMPORTANCE OF TIME SERIESS
b. It helps in planning the future, since time series can be used for prediction and
forecasting, it is one of the important tools used for planning.
Time plot is the graph of the original data over time, Chatifield emphasised that
the first step in analysis of a data is to plot the observation (Xt) against time (t),such that
the features like trend, seasonality discontinuous will be depicted if present in the
series. It is the graphical representation of time series as a plot of observations (X t)
against time so as to observe the inherent characteristics of the data.
There are four methods by which trend can be estimated. They are:
15
The least square estimates of a and b are the solution to the normal equations, which
can be determined by solving the equation simultaneously. The estimates are then given
as:
𝑎̂ = Yt- bt ( 2. 2)
Tt = 𝑎̂+ bt ( 2.3)
One advantage of this method is that a linear trend may be fitted into a series which is
not exactly linear.
To determine the seasonal factor St in a given series, we must estimate how the
data in the series vary from month to month throughout a typical year. A set of numbers
showing the relative values of a variable during the months of the year is called
seasonal index sum of 12months index number should be 1200%. There are various
methods available for computations are:
In this method, we express the data for each month as percentage of average for
the tear. The percentages for corresponding months of different years are the average,
using either a mean or median: if the mean is used, it is best to avoid any extreme
values that may occur. The resulting 12 percentages gives the seasonal index. If the
mean is not 100% i.e. if the sum is not 1200%) they should be adjusted – which is done
by multiplying them by a suitable factor.
In this method, we express the data for each month as a percentage of monthly
trend values. An appropriate average of the corresponding months then gives the
required index. As above, we adjust these if they do not average to 100%.
16
Also, in this method we compute a 12month moving average. We compute a two
month average of this 12-month moving average. The result is often called a 12-month
centered moving average.
Deseasonalized data are the data which shows how the data will look like, if there
are no seasonal fluntuations. The process of deseasonalisation of a data is by dividing
each datum in the series by the corresponding value of the seasonal index.
This is done by using residual method. T is commonly used and can be obtained
by estimating the seasonal variation and the trend from the data given.After the data
have been deseasonalised, they can be adjusted for trend simply by dividing the data by
the corresponding trend values. The process of adjusting for seasonal variation and
trend correspond to dividing adjusted for trend simply by dividing the data by the
corresponding trend values. The process of adjusting for seasonal variation and trend
correspond to dividing Yt by St, Tt, which yields Ct, It (the cyclic and irregular
variations). Once the cyclic variations have thus been isolated they can be studied in
detail. A periodicity or approximate periodicity of cycles occurs; cyclic indexes can be
much th same as seasonal indexes.
Mathematically,
Where
17
𝑬𝒕
MAE = ∑𝑵
𝒕−𝟏 | | ( 5)
𝑵
𝑬𝒕
MAE = ∑𝑵
𝒕−𝟏 | |/N (6)
𝑵
|𝑬𝒕|
PMAD = ∑𝐍𝒕−𝟏 (7)
|𝒀𝒕|
𝑬𝟐𝒕
MSE = ∑𝐍𝒕−𝟏 ( 8)
𝑵
According to chat field (1980) time series is said to be stationary if the joint
distribution of Y1,Y2,……,Yn is with a regard to a displacement in time. That is (Yt,
Yt+1,……Yt+k)=(Yt+m, Yt+m+1,…..,Yt+m+k) , where t is any point in time and k,m are
integers.
A time series (Xt) is said to be strictly stationary if for any admissible set of time
points t1,t2,t3,………tn and any constants k, the joint probability distribution of
(Yt1,Yt2,Yt3,……..Ytn)is the same as the joint probability of (Yt1+k,Yt2+k,……Ytn+k). This
is also because shifting the time origin by amount k has no effect on the joint
distribution.
18
2.6.2 WEAK STATIONARY
There are several situations responsible for stationary models. Some are mentioned
below.
a) Stable environment
b) Easily correctable trend
c) Short forecasting horizon
d) Transformable series
e) Preliminary stages of model development
19
2.8.1 Graphical Analysis
Before one pursues formal tests,it is always advisable to plot the time series under
study. Such a plot gives an initial clue about the likely nature of the time series.
2.8.2 Correlogram
𝛾𝑘
𝜌𝑘 = ⁄𝛾0
𝑐𝑜𝑣𝑎𝑟𝑖𝑛𝑐𝑒 𝑎𝑡 𝑙𝑎𝑔 𝑘
=
𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒
Note that if k=0, 𝜌𝑘 =1. Since both covariance and variance are measured in the
same units of measurement 𝜌𝑘 is unitless, or pure number. It lies between -1 and
+1,as any correlation coefficient does. If we plot 𝜌𝑘 against k, the graph we obtain
is known as the population correlogram .Since in practice we only have a
realization (i.e sample) of a stochastic process, we can onlycompute the sample
covariance at lag k,𝛾0 ,
A test of stationarity (or non stationarity) that has become widely popular over the
past several years is the unit root test.
We start with
We know that 𝜌=1,that is, in case of the unit root,the model becomes a random
walk model without drift, which is known as non_stationary stochastic process.
Therefore, regress Yt on its (one period ) lagged value Yt-1 and find out if the
20
estimated 𝜌is statistically equal to 1?. If it is, then Yt is non-stationary, This is the
general idea behind the unit root test of stationarity.
=(𝜌-1) Yt-1 + Ut
∆Yt = 𝛿Yt-1+ Ut
Models of time series data can have many forms and represent different stochastic
process. When modelling variation the level of a process, three broad classes of
practical importance are the AR models, the integrated (I) models and the MA models.
These three classes depend linearly on previous data points. Combinations of these
models produce ARMA and ARIMA models. The AR fractional integrated MA
(AFIMA) model generalized the former three.
AUTO REGRESSIVE PROCESS: A very important class of time series model is the
set of auto regressive process. Thepth order is noted as AR(P)
Yt=∑∞
𝑗=0 𝜑 ∑𝑡−𝑗 ∀ 𝜑0=1 (12)
Assuming𝜑k= 0 ∀ k > q
21
The finite approximation for G.L.P model is
Yt=∑ 𝑢𝑗 ∑ 𝑡 − 𝑗
E (Yt) = 0 ∀ t
According to Box and Jekins (1976), auto regressive integrated moving average
(ARIMA) time series models from a general class of linear models that are widely used
in modelling and forecasting time series. A time series is said to follow an ARIMA
model if it satisfies the following condition;
22
2.8.5 The Augmented Dickey-Fuller (ADF) Test
An Augmented Dickey-Fuller test is a test for a unit root in a time series sample. It is an
augmented version of the Dickey-Fuller Test for a larger and more complicated set of
time series models. The Augmented Dickey-Fuller (ADF) statistics, used in the test, is a
negative number. The more negative it is, the stronger the rejection of the hypothesis
that there is a unit root at some level of confidence.
In conducting the DF test it is assumed that the error term uiare uncorrelated. But in
case the ui are correlated. Dickey and Fuller developed a test, known as Augmented
Dickey-Fuller (ADF) test. This test is conducted by ‘augmenting’ the preceding three
equations by adding the lagged values of the dependent variable Yt. To be specific
given
∆Yt= 𝜃1+ 𝜃2 t + 𝜃Yt-1+ ut (Ytis a random walk with drift around a stochastic trend).
∆Yt-2 = ( Yt-2– Yt-3 )∆Yt-1 = ( Yt-1 – Yt-2 ) e.t.c . The number of lagged difference
terms to include is often determined empirically. The idea being to include enough
terms is serially uncorrelated. In ADF in testing whether 𝜃 = 0the ADF test follows the
same asymptotic distribution as the DF statistics, so the same critical values can be
used.
23
CHAPTER THREE
3.0 INTRODUCTION
5000000
MSD 1.85448E+12
4000000
3000000
2000000
1000000
0
1988 1992 1996 2000 2004 2008
Year
Interpretation: From the analysis above for the petroleum product data, the fitted
equation is a representation of petroleum product as a linear function of time. It follows
a linear trend.
Accuracy Measures
MAPE 8.88058E+01
MAD 1.16253E+06
MSD 1.85448E+12
5000000
MSD 1.81276E+12
4000000
3000000
2000000
1000000
0
1988 1992 1996 2000 2004 2008
Year
Interpretation: The plot above shows the quadratic trend plot analysis of petroleum
production. The fitted equation is representation of petroleum production as a quadratic
function of time.
Ŷt=14345639+285134*t-4778*t**2
Accuracy Measures
MAPE 8.75369E+01
MAD 1.10816E+06
25
MSD 1.81276E+12
5000000
MSD 2.19739E+12
4000000
3000000
2000000
1000000
0
1988 1992 1996 2000 2004 2008
Year
Interpretation: From the analysis above for the petroleum product data, the fitted
equation is a representation of petroleum product as a linear function of time. It follows
a linear trend.
Ŷt=1642124*(1.0592**t)
Accuracy Measures
MAPE 7.79831E+01
MAD 1.3352E+06
MSD 2.19739E+12
26
Interpretation: After considering the three models, the quadratic model trend gives the
best model out of all the models considered. This is due to the model which has the
least Mean Absolute Deviation.
7000000
6000000
5000000
PETROL(PMP)
4000000
3000000
2000000
1000000
0
1988 1992 1996 2000 2004 2008
Year
Interpretation: The plot above shows the time series plot of petrol. It can be deduced
from the plot that low sales of petrol is not stable over the years under consideration.
YEARS FORECAST
1998 4354667
1999 4549091
2000 4722909
2001 4876121
2002 5008727
2003 5120727
2004 5212121
2005 5282909
2006 5333091
2007 5362667
2008 5371636
2009 5360000
2010 5327758
2011 5274909
2012 5201455
2013 5107394
27
2014 4992727
2015 4857455
2016 4701576
2017 4525091
2018 4328000
2019 4110303
2020 3872000
2021 3613091
2022 3333576
2023 3033455
Interpretation: The plot above shows that there will be decrease in sales of
petroleum over the years.
28
CHAPTER FOUR
4.1 INTRODUCTION
This chapter summarizes the findings, conclusion and recommendation on the sales
of petroleum in Osun State (Oshogbo).
From the analysis, after considering the three models, the quadratic model trend
gives the best fitted model out of all the models considered. This is due to the model
which has the Least Mean Absolute Deviation.
From our forecast, it can be deduced that there would be steady decrease in sales of
petroleum in years to come.
4.4 RECOMMENDATION
Based on the analysis and the result, the following recommendations are made:
1. Governing arm of the company is advised to improve labour force from time to
time.
2. Sales of refined products should also be monitored by the Federal Government,
in order to ensure a violent free environment for activities of the company.
3. The company should make use of proposed quadratic model in this research work
to foresee future pattern of the product and to enjoy steady increase in the sales of
the product.
29
REFERENCES
Agbe Jule T.A.O.L. (1987) “Collection of revenue in Nigeria”, seminar held at the
Federal Place Hotel, Lagos.
Ailemen, M.I. and Oleosodo, L.A. (2000), The Oil Industry and Environmental
problems in Nigeria(A Case study of Nigeria Delta Area).
Anderson O.D. Time Series Analysis and Forecasting (The Box – Jenkins)
approach) the Butter Worth Group.
Augustine O.O Petroleum and the Nigeria Economy, Africa Top Publisher Ltd
Schaums outline series: statistics. New York. McGraw Hill Book Co.
Walpole, R.E (1982). Introduction to Statistics. 3rd edition New York Macmillan
Publishing Company Inc. Warren Gilchrist Statistical Forecasting a Willey –
Interscience Publications
30
APPENDIX
YEAR PETROL(PMP)
1988 1110000
1989 1080000
1990 2580000
1991 2980000
1992 3450000
1993 3950000
1994 210000
1995 4250000
1996 4450000
1997 4260000
1998 4560000
1999 4250000
2000 5120000
2001 6700000
2002 6400000
2003 2540000
2004 2950000
2005 3020000
2006 3990000
2007 4660000
2008 5320000
2009 6542000
2010 4560000
2018 7650000
31
32