Research
Article
RELATIONSHIP BETWEEN INTERNATIONAL CRUDE
OIL PRICE AND THE INFLATION RATE (CPI) IN INDIA FROM 2011 TO 2014
*B.Mahammad Rafee 1 and
Dr.A.Hidhayathulla2
1. Research Scholar (Ph.D MANF SRF), Jamal Mohamed College, Tiruchirappalli,
India
2. Associate Professor of Economics,
Jamal Mohamed College, Tiruchirappalli, India
Manuscript Info
Received:
15 March 2015
Final
Accepted: 22 April 2015
Published
Online: May 2015
Key
words:
Crude oil price CPI inflation rate
Petroleum Pricing Policy Augmented Dickey-Fuller Test (ADF) Granger Causality
test
*Corresponding
Author
B.Mahammad Rafee
Abstract
India
meets 70% of its energy needs by crude oil imports. The Price of Petroleum (per
barrel of Crude oil of 159 liters) at the international market influences the
prices of domestic petrol and diesel (domestic prices linked to International
energy derivative market). Any fluctuations in the international crude oil
price influence all other macro economic variables and Inflation too. CPI
(consumer price Index) is said to be a perfect measure of inflation by the
economist. So, for the accurate prediction of the relationship between
petroleum price and inflation, the CPI inflation is considered for the
analysis. The study proposes to use Augmented Dickey-Fuller Test (ADF) unit
root test and Granger Causality test. Crude oil price and CPI Inflation monthly
data from 2011 to 2014were used to find the exact relationship. Apart from that
the paper focuses on the petroleum pricing policy of India in brief. The study
confirms with the empirical analysis that the consumer price Inflation is not
influenced by the hike in crude oil price.
Copy Right, IJAR, 2015,. All rights reserved
INTRODUCTION
India is a crude oil importing country which meets its
70% of her energy needs through imports from Middle East and other countries
having meager percentage of gas and petroleum reserves. The price of a barrel of oil is highly dependent on both its grade,
determined by factors such as its specific gravity or API
(American Petroleum Institute) its sulphur content, and its location. Other
important benchmarks include Dubai, Tapis, and the OPEC basket.
The Energy Information Administration uses
the imported refiner acquisition cost, the weighted average cost of all oil imported into the US, as its
"world oil price”. The price of oil underwent a significant decrease after
the record peak of US$145 it reached in July 2008. On December 23, 2008, WTI
crude oil spot price fell to US$30.28 a barrel, the lowest since the financial
crisis of 2007–2010 began, and
traded at between US$35 a barrel and US$82 a barrel in 2009.On 31 January 2011,
the Brent price hit $100 a barrel for the first time since
October 2008, on concerns about the political unrest in Egypt. Recently the price of
crude oil dips to the lowest at $49 a barrel since 2009 it’s the lowest. Any
fluctuation in the price affects the growth, inflation, forex reserves and
widens the CAD (Current Account Deficit) of crude importing countries. Few
countries china and Malaysia are able to maintain the fixed price for crude oil
and India’s inability to maintain the fixed pricing policy for petroleum costs
the petroleum products users severely. Petrol and diesel are the major fuels
for transportation, Industrial and other purposes. Fluctuations in the price of
petrol and diesel have a cascading effect on Indian economy. The study throws a
light on the relationship between Crude oil price and the Inflation rate only.
The paper further divided in to four sections. The second section dedicated to
Pricing of petroleum in India. The third section gives the literature review on
impact of crude oil price on CPI Inflation. The fourth section dedicated to
methodology and empirical analysis. Finally the fifth section offers concluding
comments.
2. Pricing
of Petroleum in India
Crude
oil, both indigenous and imported are refined in to various petroleum products
viz., petrol (motor spirit), naphtha, light diesel, aviation fuel, kerosene,
high speed diesel, furnace oil, bitumen, waxes etc. The pricing of refined
petroleum products have gone through various phases beginning from value stock
accounting system and import parity pricing and then to retention pricing under
Administrated price mechanism (APM) and presently trade parity pricing. Till
1939, there was no control on the pricing of petroleum products. Between, 1939
to 1948 the oil companies themselves used to pool accounts for major products
without intervention of the government. However, since independence the pricing
of petroleum products witnessed several structural changes in policies. In
1948, an attempt was made to regulate prices through valued stock account
procedure. This was a cost plus formula based on import parity to which
additions like ocean freight up to Indian ports, insurance, ocean loss,
remuneration, import duty and other levies and changes. The realization of oil
companies under this procedure was restricted to import parity price of
finished goods plus excise duties/local taxes/dealer margins and agreed
marketing margins of each of the refineries. Any realization in excess of
normal was surrendered to the government.
The
petroleum industry has been deregulated with the intention of shifting to market
determined pricing mechanism. However in practice the deregulation process has
been only implemented partially due to restriction on prices imposed by the
Government to shield the Indian consumer from oil price volatility especially
since 2004.
The
process of deregulation of petroleum product prices begun in 1998, five
sensitive products namely petrol, diesel, domestic LPG, PDS kerosene, ATF
(Aviation Turbine Fuel) continued as controlled commodities. In the post-APM
era beginning from 1-4-2002, oil marketing companies were allowed to sell their
products at market determined prices. It is based on the notion of import
parity from April 2002 to May 2006 and from June 2006 on wards on the basis of
trade parity for petrol and diesel (except PDS Kerosene and LPG subsidy
continued) after consultation with the ministry of petroleum and natural gas
(MoPNG).In 2004 the prices started rising in the international market. Although
the oil marketing companies were granted freedom to fix retail selling price
fortnightly basis, price used to be
revised after informal clearance from MoPNG and there was no price revision of
petrol and diesel during the period of mid 2004 ruling price international
market were abnormally high during this period.
In
August 2004 Government worked out new methodology allowing OMC’s limited
freedom to revise the prices of Petrol/diesel within a price band. The concept
of price band on the principle of rolling average price of these products in
the International market accordingly, oil companies were permitted to carryout
autonomous adjustments in prices within a band of +/- 10% of the mean of
rolling average CIF price of preceding 12 months and preceding quarter i.e.
three months. In case of breach of this band, the OMC’s had to approach the Ministry
of Finance (MoF) through ministry of Petroleum and Natural gas (MoPNG) to
modulate the excise duty rates so that the spiraling process prevailing in the
international markets do not cause undue hardship to the consumer. However
consequent to the further rise in oil prices, price band approach had to be
given up.
Oil refining and marketing companies become all the
more worse due to high volatility of oil prices and they have made huge losses.
The oil companies reported financial distress in terms of “Under recoveries”,
with respect to import parity price formula that has been in use since the end
of the APM regime. A separate section has been devoted in the post-APM era
between 2003-2008 on account of asymmetric price adjustments between
international crude oil prices, domestic prices of sensitive petroleum products
i.e. High powered committee on financial position of oil companies. Presently
trade parity pricing has been in practice for petroleum products for refinery
gate as well as retail pricing (recommended by Rangarajan committee) and
proposed to review and update the trade parity price every year depending on
the relative weights of exports and imports.
Based on the
recommendations of the Kirit Parikh Committee, the Government of India (GOI) on
25 June, 2010 announced the full deregulation of the prices of two crucial
petroleum products: petrol and diesel. Henceforth, prices of these two products
will be determined by the unfettered play of market forces and government
“subsidies” on these products, which worsen the fiscal situation, will be
completely removed. Government control over the determination of the prices of
these key commodities was willingly ceded to the magic of the market,
presumably to “rationalize” prices and to wipe away losses of state-run Oil
Market Companies (OMCs) to the tune of Rs. 22,000 crore.
The markets were
ecstatic about the full liberalization of petrol and diesel prices and these
sentiments were almost immediately reflected in rising oil stock prices. There
were strident complaints that this policy change was not enough: prices of
kerosene and liquefied petroleum gas (LPG) were still minimally under
government control and therefore even after the deregulation move, the losses
of the OMCs on account of these two petroleum products would stand at Rs.
53,000 crore for fiscal 2011.The first crucial victory of this struggle came in
2002 when the government dismantled the administrative pricing mechanism (APM).
This move reduced the “subsidies” on petrol and diesel but the government
decided to continue to “subsidize” kerosene and LPG. In 2005, the GOI
constituted the Rangarajan Committee to study pricing and taxation of petroleum
products. This committee recommended a half-way house: a ceiling on the
refinery gate price (computed according to the so-called trade parity formula)
along with the freedom for OMCs to set retail prices. Of course, this was not
enough. Accordingly, in 2009 the next committee was constituted to examine the
same set of issues, i.e., the Kirit Parikh Committee. In its report submitted
in February 2010, the Kirit Parikh Committee finally recommended what the
capitalist sector had been telling GOI all these years. It recommended full
liberalization of petrol and diesel prices.
The new government in 2014 has deregulated the price of diesel too as
according to the energy policy.
3. Literature review
Food
sector prices are influenced by high speed diesel oil prices as diesel is fuel
for trucks to carry the agricultural output from one part of the country to other.
Syed Atif Ali et.al (2012) examined the effects of high speed diesel oil prices
on food sector prices in Pakistan using multiple linear regression. The food
includes rice, Maize, wheat, chicken and cooking oil which are dependent
variables in the study. The independent variables are high speed diesel. It is
hypothesized that there is a significant relationship and positive effect of
oil prices on food inflation. The study concludes with a support of hypothesis
that there is a highly significant effect of oil prices on food inflation.
(High speed diesel price found to be have highly significant effect on food
inflation in Pakistan)
Oil
price shocks have a sudden transmission in the economies through inflation.
Benjamin Wong (2012) found the impact of different oil shocks on US inflation
and inflation expectations since 1970’s. The findings confirm oil supply shocks
have never been a major factor, demand side shocks in the oil market generally
been more important in explaining inflation dynamics and movements, inflation
expectations. The authors said that exogenous political events induce oil
shocks that are more inflationary. The author concludes that demand shocks in
the oil market have a much larger role for inflation and inflation
expectations. The response to oil supply shocks that raise real oil prices by
the same magnitude doesn’t appear to exhibit time variation invoking a
hypothetical thought experiment where demand side shocks in the oil market
raise the real oil price by a fixed magnitude (say 10%) shows a large drop off
in the response of inflation and inflation expectations. (Exogenous political
factors induce oil shocks that are more inflationary and demand shocks in the
oil market have a much larger role for inflation and inflation expectations
than supply shocks).
Consumer
price Index (CPI) is a best indicator of inflation than Whole sale price index
(WPI). Surjit Bhalla (2011) studied that across most countries emerging and
developed, the best indicator of overall inflation (as measured by GDP
deflator) is the consumer price index (CPI). Policy makers in India, including
the RBI have been erroneously using the whole sale price index (WPI) as a
surrogate for underlying inflation even when its ability to accurately forecast
overall inflation is close to zero, especially in the presence of information
on CPI inflation. Since Feb. 20th 2011, a new national CPI index has
been released with urban and rural all India components. Indian inflation for
the last thirty years is strongly correlated to international inflation which
in turn is correlated to commodity prices over which domestic monetary policy
has little control, each $10 rise in oil price increases inflation by about
0.5% for emerging markets, including India. For developed economies, the effects
are muted- each $10 in the price of oil raises the inflation rate by only 0.03
percent. (Crude oil price rise highly correlates with CPI index and confirms
domestic monetary policy has a little control).
With
the above literature background the study analyses the relationship between the
Crude oil price and Consumer Price Index (Inflation) from January 2011 to
September 2014.
Material and Methods
The data on oil prices were downloaded from eia.gov.
Data relating to consumer price index is downloaded from labour bureau of
India. The study employs an empirical analysis and only focuses on the two
chosen variables. The variables that we use are the world crude oil prices in
US Dollars, Consumer price inflation rate. Time series data from January 2011
to September 2014 are used for the variables.
Statement of Hypothesis
The hypothesis for this study has been stated below
Null Hypothesis:
H0:
There is no significant relationship between crude oil price and
inflation rate (CPI)
H1:
There is a significant relationship between crude oil price and
inflation rate (CPI)
Descriptive
statistics technique
Descriptive
statistics is the discipline of quantitatively describing the patterns and
general trends of a dataset and summarize it in single value. It enables a
reader to quickly understand and interpret the set of data that has been
collected. In this study, descriptive statistics provide a useful quantitative
summary of the variables. Here, descriptive statistics provide a historical
account of variables behavior and convey some future
aspects of the distribution of dataset. The study used measures of central
tendency (mean) and measures of Variability (standard deviation, minimum and
maximum) to explain the dataset.
Inferential
statistics technique
Inferential
statistics is defined as the branch of statistics that is used to make
inferences/ valid judgments about the characteristics of a population based on
sample data. These statistics are ways of analyzing data that allow the
researcher to make conclusions about whether a hypothesis was supported by the
results.
A hypothesis is an educated guess about a trend, group
difference or association believed to exist. A null hypothesis states that the
results will be due to chance whereas an alternate hypothesis tells that the
results are due to the manipulation of the independent variable. Here in our
study, null hypothesis (H0) There is a significant
relationship between crude oil price and inflation rate (CPI), while alternate
hypothesis (Ha) is there is a significant
relationship between crude oil price and inflation rate (CPI).
There
are different ways to inference the results. Here, we used correlation matrix
analysis and linear regression analysis (t-ratio, f-sign, p- value, r-square)
which allows us to make a conclusion related to our hypothesis. We have used 5%
of level of significance and two tailed test so as to accept or reject our null
hypothesis according. Regression analyses are typically done using statistics
software and here we used Eviews8.
Econometric
Regression Model
The
term regression was introduced by Francis Galton. Linear
regression analysis is an inferential statistical technique that is used to
learn more about the relationship between a independent variable (referred to
as X) and dependent variable (referred to as Y) When there is only one
independent variable, the prediction method is called simple regression. So,
the regression equation Yi = β0 + β1 Xi + ui
where Yi is the dependent variable, Xi is the independent variable,
β0 is the constant (or intercept), β1 is the slope of the
regression line which represent the strength and direction of the relationship
between the independent and dependent variables and ui is random error
term. Here, in the study carried out this method to see and interpret the
effect of crude oil price on inflation rate (CPI).
Statistic test
R-square:
also known as the coefficient of determination is commonly used to evaluate the
model fit of a regression equation. That is, how good are all of your
independent variables at predicting your dependent variable? The value of
R-square ranges from 0.0 to 1.0 and can be multiplied by 100 to obtain a
percentage of variance explained.
Sign-F:
whether the model as a whole is significant. It tests whether R- square is
significantly different from zero.
T-ratios: the
reliability of our estimate of the individual beta. For that we can look at p-
values.
Unit root test
(Augmented Dickey –Fuller test)
The
stationarity of a data series is a prerequisite for drawing meaningful
inferences in a time series analysis and to enhance the accuracy and
reliability of the models constructed. If the variable is not stationary
estimation can obtain a very high R2, although there is no
meaningful relationship between the variables. This situation reflects the
problem of spurious regression between totally unrelated variables generated by
a non-stationary process. Generally a data series is called a stationary series
if its mean and variance are constant over a given period of time and the
covariance between the two extreme time periods does not depend on the actual
time at which it is computed but it depends only on lag amidst the two extreme
time periods.
One
of the common method is to find whether a time series is stationary or not is
the unit root test. There are numerous unit root tests. One of the most popular
among them is the Augmented Dickey-Fuller (ADF) test. Augmented Dickey -Fuller
(ADF) is an extension of Dickey -Fuller test. Following equation of ADF test
checks the stationarity of time series data:
where
Yt is the variable in period t, T denotes a time trend,
is the difference operator, et is an error
term disturbance with mean zero and variance σ2 , and k represents
the number of lags of the differences in the ADF equation. The ADF is
restricted by its number of lags. It decreases the power of the test to reject
the null of a unit root, because the increased number of lags necessitates the
estimation of additional parameters and a loss of degree of freedom. The test
for a unit root is conducted on the coefficient of yt-1 in the
regression. If the coefficient is significantly different from zero (less than
zero) then the hypothesis that y contains a unit root is rejected. Rejection of
the null hypothesis denotes stationarity in the series.
Null and alternative
hypothesis are as follows:
|
|
H0 : ρ=0
|
[Variable
is not stationary]
|
Ha :
ρ<0
|
[Variable is
stationary]
|
Our study also contains time series data. The time
series variables considered in this paper are the stock market indices and
seven macroeconomic variables. This necessitates the inclusion of ADF test in
the present study. Also our study includes Granger causality test which assumes
that the variables involved are stationary. Thus prior to testing and
implementing the Granger Causality test, econometric methodology needs to
examine the stationarity for each individual time series. If the variables are
not stationary the standard assumptions for asymptotic analysis in the Granger
test will not be valid.
Null hypothesis in this case would be that particular
CPI Inflation rate and Crude oil price is not stationary & alternative
being that they are stationary.
Note: We
have considered p- value for testing at 5% significance level. If the p-value
is smaller than 0.05 then Null hypothesis will be rejected &
variables would be stationary & vice versa.
ADF test is used to find
the stationarity or non-stationarity of the variables in the data series.
Inferential statistics techniques are used to inference about the results like
multiple linear techniques after attaining stationarity of both the series
using E-VIEWS8.0.
Granger causality test
Granger
(1969) and Sim (1972) were the ones who first developed Granger causality test
to examine the application of causality in economics. Granger causality test is
a technique for determining whether one time series is significant in
forecasting another. The standard Granger causality test seeks to determine whether
past values of a variable helps to predict changes in another variable. Granger
causality technique measures the information given by one variable in
explaining the latest value of another variable. In addition, it also says that
variable Y is Granger caused by variable X if variable X assists in predicting
the value of variable Y. If this is the case, it means that the lagged values
of variable X are statistically significant in explaining variable Y. The null
hypothesis (H0) that we test in this case is that the X variable
does not Granger cause variable Y and variable Y does not Granger cause
variable X. In summary, one variable (Xt) is said to granger cause
another variable (Yt) if the lagged values of Xt can
predict Yt and vice-versa. The test is based on the following
regressions:
Where Yt and Xt are the
variables to be tested, and ut and vt are mutually
uncorrelated errors, and t denotes the time period and ‘k’ and ‘l’ are the
number of lags.
The null hypothesis is:
H0 : αt
= δt = 0 for all i
|
[X does not granger
cause Y]
|
The alternative
hypothesis is:
|
|
Ha : αt
≠ 0 and δt ≠ 0 for at least
some i
|
[X granger cause Y]
|
If the coefficient αt
are statistically significant but δt are not, then X causes Y. In the reverse case,
Y causes X. But if both αt
& δt are significant, then causality runs both ways. The null
hypothesis is tested by using the standard F-test of joint significance. The
F-test is applied, as follows:
F = (RSSR – RSSUR)/m
RSSUR/
(n-k)
Here RSSR & RSSUR are the
restricted and unrestricted residual sum of squares respectively. M is the
number of lags, n is the number of observations and k is the parameters in the
unrestricted equation. If the computed F-value exceeds the critical F-value
at the chosen level of significance, the null hypothesis is rejected. This
would imply that macroeconomic variable ‘Granger cause’ or improve the
prediction in stock prices and vice versa.
Note: That
it has been taken one period lag in the above equation. In practice, the choice
of the lag is arbitrary.
In
the present study Granger Causality Model has been used to test the causality
between Crude oil price and CPI inflation rate. Here the test signifies whether
past information on macroeconomic variables predict stock prices in India, Null
& Alternative hypothesis being:
H0:
There is no significant relationship between crude oil price and
inflation rate (CPI)
H1:
There is a significant relationship between crude oil price and
inflation rate (CPI)
Note: A lag of five months has been considered.
Empirical analysis:
Crude oil is an indispensable input for production and
therefore, the price of oil is included as a proxy for real economic activity.
India is largely an importer of crude oil and consequently, oil price takes
part an imperative role in Indian economy. It is apparent that any key movement
in oil prices leads to uncertainties in the stock market which could persuade
investors to suspend or delay their investments.
Moreover,
increase in oil prices results in higher transportation, production and heating
costs which have negative effect on corporate earnings. Rising fuel prices also
raise alarm about inflation and diminish consumers’ discretionary spending. Therefore,
the financial risk of investments increases when there is wide fluctuation in
oil prices. Therefore, for oil importing
countries like India, an increase in oil price will lead to an increase in
production costs and hence to decreased future cash flow, leading to a negative
impact on the stock market. Therefore, an increase in the price of oil in the
international market means lower real economic activity in all sectors which
will cause stock price to fall.
The mean of crude oil price is 109.72 and its maximum
value is 125.45 and minimum value is 95.16, while standard deviation is 6.27.
The crude oil price data found tobe stationary at the level itself the
following graph confirms the stationarity of the data series. Apart from that
ADF test applied to know the stationarity.
ADF test results shows that the p-value is 0.001 which is less than the critical value. Where oil price is series
find tobe stationary at the level itself.
Consumer Price
Index (CPI)
Inflation
is measured by changes in the Consumer Price Index (CPI). High rate of
inflation increase the cost of living and a shift of resources from investments
to consumption. This leads to a fall in demand for market instruments which
lead to reduction in the volume of stock traded.. High rate of inflation
increase the cost of living and a shift of resources from investments to
consumption. This leads to a fall in demand for market instruments which lead
to reduction in the volume of stock traded. Also the monetary policy responds
to the increase in the rate of inflation with economic tightening policies.
Inflation is ultimately translated into nominal interest rate and an increase in
nominal interest rates increase discount rate which results in reduction of
present value of cash flows. High Inflation affects corporate profits, which in
turn causes dividends to diminish thereby lower stock prices. When inflation
begins to move upward, it likely leads to tight monetary policies which result
in increase in the discount rate. It indicates that the cost of borrowing
increases which in turn leads to investment reduction in the stock market. So,
it is said that an increase in inflation is negatively related to equity
prices.
The mean of the inflation rate is 9.12 and
the maximum and minimum rate about 12.06 and 5.3 respectively. While std.
deviation is 1.71. The ADF Unit root test states that the series is non-stationary
with the p-value of 0.2416 which is higher than the critical value and the
series find tobe stationary at the first difference.
The ADF test confirms the series stationary at the first difference
with p-value of 0.001 which is less the critical value.
Granger causality test
Granger causality test is a technique for determining
whether one time series is significant in forecasting another or not. Here
Granger-causality test has been conducted to study the causal relationship
between Crude oil price and CPI Inflation rate. The tables below reports
granger causality test results with lag of 5 that is the appropriate selection
of lags. The null hypothesis has been tested on the basis of the P-value. If
the P-value is less than the critical P value at 5% than the null hypothesis is
rejected and there will be a significant relation between the variables. First
differencing of the variables has been used to apply granger causality test.
Granger Causality Tests for Crude oil Price and CPI
Inflation Rate
Null Hypothesis
|
P-Value
|
Result
|
Relationship
|
Oil price does not granger cause CPI
rate
|
0.9369
|
ACCEPT NO RELATION
|
|
CPI rate does not granger cause oil price 0.2503 ACCEPT
The above table shows
granger causality test for Crude oil price and CPI inflation rate, the test
confirms there in no relation between crude oil price and CPI inflation rate.
The monthly data analysis of the two variables does not found the relationship.
Conclusion
The
paper examined the effects of hike in crude oil price on the domestic inflation
(CPI) rate using time series monthly data from January 2011 to September 2014.
Augmented Dickey fuller unit root test (ADF) has been used to find the
stationarity of the data series, where crude oil price found to be stationary
at the level and CPI inflation rate was non stationary at the level and brought
to stationary at the first difference. The correlation matrix and regression
does not find the perfect relationship between the two variables and Granger
causality test also confirms that there is no relation. So, the paper concludes
that the hike in crude oil price doesnot influence the domestic inflation rate as
suggested by monthly data of two variables. Daily and weekly data of crude oil
price and weekly inflation rate can be verified to know exact relationship.
References
1. Syed Atif Ali
et.al (2012), “Impact of oil prices on food inflation in Pakistan”.2012.
2. Benjamin
wong (2012) “inflation dynamics, inflation expectations and the role of oil
shocks: what has changed since the 1970’s- Centre for macro economic analysis,
the Australian national university, Canberra ACT0200, Australia. May 2012.
3. Surjit Bhalla
(2011) “Developing trends”.vol.I Issue-1 March 2011.
4. Ine Van
Robays (2012) “macro economic uncertainty and the impact of oil shocks”.
European central Bank working paper
series: 1479/October 2012. ISSN 1725-2806 (online).
5.
Kaushik Rajan Bandyopadhyay
(2009). “Petroleum Pricing in India: Transition from APM to MDPM”, Asian
Institute of transport development-MPRA Paper No: 25905/2009.
6. Surya Sethi (2010) The political
Economy of oil prices in India: EPW article: “Analysing the
Parikh Committee Report on Pricing of Petroleum Products,” Economic and Political Weekly,
March 27, 2010.
7. Dickey, D.A, Fuller, W.A. (1979). “Distributions of
the Estimators for Autoregressive Time Series with a Unit Root”, Jornal of
American Statistical Association 74:366, pp. 427-481.
8. Dickey, D.A. and W.A. Fuller. (1981). Likelihood Ratio
Statistics for Autoregressive time Series with a Unit Root, Econometrica 49,
1057-1072.
9. Granger, C.W.J, (1969), Investigating Casual relations
by Econometric Models and Cross-Spectral Models, Econometrica 37, 428-438.
10.
Gujrati, D.N. (2003). Basic Econometrics,
McGraw Hill, India.
