See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/320831652

ANALYSIS OF AGRICULTURAL TRADE BETWEEN PAKISTAN AND UNITED ARAB

EMIRATES: AN APPLICATION OF GRAVITY MODEL

Article in Journal of agricultural research · January 2016

CITATIONS READS

6 441

4 authors, including:

Iqbal Javed Muhammad Ashfaq

University of Lahore Sargodha Campus University of Agriculture Faisalabad
29 PUBLICATIONS 53 CITATIONS 138 PUBLICATIONS 1,179 CITATIONS

SEE PROFILE SEE PROFILE

Bakhsh Khuda
COMSATS University Islamabad, Vehari campus, Pakistan
86 PUBLICATIONS 910 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Factors effect migration intention among youth View project

AgMIP Phase- View project

All content following this page was uploaded by Iqbal Javed on 03 November 2017.

The user has requested enhancement of the downloaded file.

 J Agric. Res., 2016, Vol. 54(4):787-799
ISSN: Online: 2076-7897, Print:0368-1157
http://www.jar.com.pk

ANALYSIS OF AGRICULTURAL TRADE BETWEEN

PAKISTAN AND UNITED ARAB EMIRATES:
AN APPLICATION OF GRAVITY MODEL
Iqbal Javed, Muhammad Ashfaq, Sultan Ali Adil and Khuda Bakhsh*

ABSTRACT
This study was carried out at the Institute of Agricultural and Resource
Economics, Faculty of Social Sciences, University of Agriculture, Faisalabad,
Pakistan during 2012-2013. The objective was to analyze the major factors
affecting the trade between Pakistan and United Arab Emirates (UAE). Impact
of different variables was determined by the application of gravity model using
panel data. Variables that were used are total trade, population, GDP, distance
between trading partners and dummies for border and cultural similarities.
Fixed effect model was selected for the time variant variables while random
effects model was selected for time invariant variable using F-test, Housman
specification test and Breusch-Pagan Lagrange multiplies test. The results
showed that value of coefficient of GDP of Pakistan was 0.75 which was
significant. The value of coefficient of GDP of UAE was 0.69 which was also
significant. The coefficient of population of UAE was 0.63 which was
significant. The estimated value of coefficient of population of Pakistan was -
1.24 by fixed effect model. The similar culture of Pakistan with its major trading
partners have 1.27 percent more trade as compared to trading partners which
have no cultural similarities with Pakistan. The joint border countries of
Pakistan have 2.3 percent less agricultural trade as compared to the countries
which have no joint border. Distance between trading partners has a negative
impact on agricultural trade but this was not significant.

KEYWORDS: Agricultural trade; gravity model; fixed effect model; United Arab
Emirates; Pakistan.

INTRODUCTION

Agriculture has a very important role in economy of Pakistan with a share of

about 21.4 percent to GDP. It provides 45 percent employment to the total
labor force of the country (2). About 60 percent rural populations depend on
agriculture (1). Pakistan’s export value was US$ 20,147 million for ten

*Institute of Agricultural and Resource Economics, Faculty of Social Sciences,

University of Agriculture, Faisalabad, Pakistan.
788 I. Javed et al.

months from July to April during fiscal year 2012-13 (2). The imports of
Pakistan amounted to about US$ 44,912 million during 2012-13 (3). Pakistan
generally has a negative trade balance. Major trading partners of Pakistan
are China, Saudi Arabia, United Arab Emirates (UAE), United States,
European Union Kuwait, India and Malaysia. Pakistan has about 17 percent
of total trade with China. Trade share of Pakistan with UAE was about 11
percent of its total trade with an export share of 8.5 percent and import share
of 12.3 percent. Pakistan has a trade share of 9 percent with Saudi Arabia.
Trade share of Pakistan with European Union is 13.0 percent. Trade flow
between Pakistan and United States has been decreasing since last few
years and in 2012-13 it was only 6.7 percent with exports (13.3 percent)
exceeding the imports (3.2 percent). The other countries like Kuwait, India
and Malaysia have a minor trade share with Pakistan which is 4.4, 3.2 and
2.9 percent respectively (3).

Pakistan’s major export items to United Arab Emirates include clothing of

textile fabrics, hosiery, rice, cotton fabrics, cotton yarn, sports goods, fruits,
vegetables and footwear. Major imports of Pakistan from UAE are petroleum
products, precious stones, metals, plastic products, iron, steel, machinery,
organic chemicals and electrical equipment. Pakistan has been exporting
agricultural products and importing non-agricultural products from United
Arab Emirates.

A set of opportunities always exist in bilateral trade. The approach made

analyses of trade of Pakistan with UAE No such detailed study was found for
a special market that have a large share in Pakistani agricultural trade so the
present study was planned to analyze the impact of major variables on trade
and to make analysis of trade for a particular market of UAE as a trading
partners.

A study was carried out on bilateral trade between Pakistan and turkey by
gravity model using the panel data set of many trading partners (18).
Mohmand and Wang (15) described that trade environment of Pakistan was
facing chronic trade deficit by using the gravity model. Zarzoso (21) applied
the gravity model of trade to determine Mercosur-European Union trade, and
potential of the trade following the agreements made recently between these
both trade blocs. Rahman (16) carried out research on panel data analysis
for trade of Bangladesh using gravity model approach’. He provided
theoretical justifications for using the gravity model in bilateral trade analysis.
Roy and Rayhan (17) conducted a study about import flows of Bangladesh
using gravity model approach under Panel Data Methodology. Iqbal (11)

J. Agric. Res., 2016, 54(4)

 Agricultural trade between Pakistan and UAE by using gravity model 789

used gravity model and found that European Union was the most important
destination of exports of Bangladesh.

MATERIALS AND METHODS

This study was conducted at the Institute of Agricultural and Resource

Economics, Faculty of Social Sciences, University of Agriculture, Faisalabad,
Pakistan during 2012-2013.

Data collection, variables, and sources: For the determinants of

agricultural trade between Pakistan and United Arab Emirates different
variables were selected. Panel data were used for the analysis for a duration
of 1975 to 2012. The data about indicators were taken from the World Bank,
Federal Bureau of Statistics, Islamabad, Pakistan, State Bank of Pakistan,
Pakistan Economic Survey, International Financial Statistics and International
Monetary Fund. Variables used were, GDP of Pakistan, GDP of major trading
partners, population of Pakistan, population of major trading partners,
distance between Pakistan and its major trading partners. Dummy variables
for border and cultural similarities were used in the models. To determine the
impacts of different variables on agricultural trade of Pakistan with UAE as a
major trading partner, the proxy of total volume of bilateral trade was used as
a dependent variable in the model because the time series data of bilateral
agricultural trade were not available. Pakistan’s major portion of trade is
agriculture, and factors which determine the overall trade also affect the
agricultural trade.

Testing for panel unit root: To check the existence of unit root in panel
data, different tests were used. These are Levin, Lin & Chu (LLC), I P, Shin
W-stat, ADF - Fisher Chi-square and PP - Fisher Chi-square. If hypothesis of
unit root is not rejected, then first difference was tested for presence of unit
root and so on. This procedure continues until the null of unit root is rejected.

Gravity model of trade: The gravity model about the trade was first used by
Tinbergen in 1962 (19). The basic model for trade between two countries (i
and j) takes the form of:

Here F is the trade flow, M is the economic mass of each country, D is the
distance and G is a constant.

J. Agric. Res., 2016, 54(4)

790 I. Javed et al.

Econometric estimation of gravity equations: Using panel data set, three

gravity models were estimated for analysis of bilateral trade between
Pakistan and UAE. For the model work of earlier scientists (15, 17) was
followed. Rahman carried out a study on trade of Bangladesh by using
gravity model Approach (16). The gravity model of trade in this study is:
log (TRijt) = β0 + β1 log (GDPit) + β2 log (GDPjt) + β3 log (Populationit) + β4 log
(Populationjt) + β5 log (Distanceijt) + β6 (Borderij) + β7 (Cultureij) + Uijt

where,
i = Pakistan
j = Trading partner of Pakistan
TRij = Total trade between country i and country j,
GDPi (GDPj) = Gross Domestic Product of country i and (j),
POPi (POPj) = Population of Country i and (j),
Distanceij = Distance between country i and country j,
Cultureij = Cultural similarities between both trading partners i and j
(dummy variable),
Borderij = Land border between country i and j (dummy variable),
Uij = error term; t = time period, βs = parameters.

RESULTS AND DISCUSSION

For the analysis by gravity model, panel data were used in determination of
agricultural trade of Pakistan with United Arab Emirates. Variable of inflation
rates and per capita income were dropped from the model due to problem of
multicollinearity.

Stationarity Tests: Presence of a mutual trend between any two time series
data does not always infer that there is a important economic association
between them. If the time series were not stationary, the regressions
containing these time series can incorrectly imply the presence of
association. That is called spurious regression (8). Ignoring these facts and
estimating a regression model comprising non-stationary variables might lead
to incorrect results. To check the existence of unit root in panel data different
tests were used. These tests for unit root of panel data are Levin, Lin & Chu,
I P, Shin W-stat, ADF - Fisher Chi-square and PP - Fisher Chi-square and
results are given in Table 1.

Test statistics of four methods used (Levin Lin & Chu, I P Shin W-stat, ADF -
Fisher Chi-square and PP - Fisher Chi-square) for GDPi at level form are not
significant which indicated the data as non-stationary at level form. After

J. Agric. Res., 2016, 54(4)

 Agricultural trade between Pakistan and UAE by using gravity model 791

transferring the data of GDPi in first difference form the test statistics of all
four methods became significant. It indicated that the data of GDPi after first
difference, turned to stationary. Test statistics of four methods used for GDPj
at level form are not significant. It indicated that data was non stationary at
level form. After transferring the data of GDPj in first difference form the test
statistics of all four methods became significant, which indicated that data of
GDPj after first difference, became stationary. The value of four tests for the
data about POPi and POPj are significant at level form, indicating that the
data was stationary at level form.
Table 1. Results of Panel Unit Methods.

Variables Data Type/probability Levin, IP & ADF - PP - Fisher

Lin & Shin Fisher Chi-square
Chu t* W-stat Chi-square
GDPi Probability of Level data 0.22 0.99 1.00 1.00
Probability of 1st Difference 0.000 0.000 0.000 0.000
GDPj Probability of Level data 0.126 0.993 0.000 0.000
Probability of 1st Difference 0.000 0.000 0.000 0.000
Probability of Level data 0.090 0.79 0.67 0.67
Lbtrade
Probability of 1st Difference 0.000 0.000 0.000 0.000
POPi Probability of Level data 0.000 0.000 0.000 0.000
POPj Probability of Level data 0.000 0.002 0.000 0.000
Source: Author’s calculations
Determinants of bilateral trade by using the gravity model: Panel data
might provide additional advantages, capturing relationship over variables in
time and observing individual effects between trading partners (4, 12). Fixed
effects model (FEM) would be a better choice than random effect model
when one is interested in estimating trade flows between a predetermined
selections of nations (7). The problem faced by FEM is that it cannot directly
estimates variables that do not change over time, such as distance, border,
and cultural similarities, because the inherent transformation tends to wipe
out such variables. For the model the authors followed Rahman (16) and
Suvankulov and Ali (18). In that context, an effort was made to quantify the
impact of major determinants on trade of Pakistan with United Arab Emirates
as a major trading partner. Gravity model of trade with Double log form of
model with first differences was used to find the impact of various factors on
trade between Pakistan and United Arab Emirates.
Descriptive statistics (minimum, maximum, mean, and std. deviation) were
used to describe the data of dependent (bilateral trade between Pakistan and
its major trading partners) and independent variables (GDP, population,
distance, dummy variable of border and dummy variable of cultural
similarities) as shown in Table 2.

J. Agric. Res., 2016, 54(4)

792 I. Javed et al.

Table 2. Summary statistics of panel data used for the gravity model.

Variable N Minimum Maximum Mean Std.

Deviation
GDPI (Billion US$) 380 11.34 231.20 70.19 55.82
GDPJ (Billion US$) 380 7.85 16244.6 1517.69 2885.00
POPJ (Million) 380 0.53 1350.69 263.62 408.70
POPI (Million) 380 68.48 180.00 123.04 34.14
DISIJ (KM) 380 686.00 10866.00 3967.90 2851.59
BTRADE 380 15.65 10859.97 1267.22 1635.88
Source: Author’s calculations

Selection of model: We have paid particular attention to the fact that gravity
model not only contain time varying variables such as GDPs and population
but also includes time invariant variables namely distance, border and culture
similarities. Fixed effect model does not allow for estimating time invariant
variables (16). However, pooled OLS and random effects treatment allows
the model to contain observed time invariant characteristics. Hausman
specification test was used to check whether the fixed effect or the random
effect model (REM) was appropriate. F-test was used to choose one best
model from the model of fixed effects and pooled OLS. Breusch-Pagan LM
test was used for the selection of appropriate model between pooled OLS
and random effect model.

Hausman specification test (FEM or REM): Hausman specification test is a

common method to compare the fixed and random effects estimates of
coefficients (5, 10, 20). To choose FEM or REM, Hausman test was used
which has an asymptotic chi-square distribution.

P-value = prob (chi-square (4) >61.412759) = 0.0000

Small p-value, by Hausman test was in favor of fixed effects.

F-Test (FEM or Pooled OLS): To check which model is better, a formal test
for two models was used. Pooled OLS model is used as the baseline for our
comparison. Highly significant value of F-test (F=43.95) counted against the
null hypothesis that the Pooled OLS model was adequate, in favor of the
fixed effects alternative.

Bruesch-Pagan LM Test (Pooed OLS or REM): After application of F-test

and Hausman specification test, it was concluded that FEM was appropriate
instead of pooled OLS and random effects model could be used. Now for the
time invariant variables we could use both Pooled OLS and random effect

J. Agric. Res., 2016, 54(4)

 Agricultural trade between Pakistan and UAE by using gravity model 793

model but which model was most appropriate for the analysis was known by
the Breusch-Pagan Langrang Multiplier test (3). Significant value showed that
random effects model was appropriate instead of pooled OLS model.

Fixed Effects Model: On the basis of Hausman test and F-test the fixed
effects model was selected and the estimated gravity model (Table 3). In
fixed effects model, only time variant variables were given because time
invariant variables were not used in the fixed effect model. Zarzoso (21)
applied the gravity model of trade to determine Mercosur-European Union
trade, and potential of the trade following the agreements made recently
between these both trade blocs and He found that fixed effect model was
more appropriate as compared to random effects gravity model.

Table 3. Estimated gravity model for bilateral trade of pakistan by fixed effects.

Variable Coefficient Std. Error t-Statistic Prob.

C 2.92 1.04 2.78 0.0056***
DLGDPJ 0.69 0.01 44.05 0.0000***
DLGDPI 0.75 0.12 6.00 0.0000***
LPOPI -1.24 0.31 -3.93 0.0001***
LPOPJ 0.63 0.01 42.31 0.0000***
Adjusted R-squared 0.90
Durbin-Watson stat 2.08
F-statistic 263.63
Prob (F-statistic) 0.000
Adjusted R-squared 0.90, Durbin-Watson stat 2.08, F-statistic 263.63, Prob (F-Statistic), HAC
standard errors are used for removal of heteroscedastisity, * = (P< 0.05), *** = highly
significance, NS = Non Significant

According to F test and Housman specification test, fixed effect model was
considered as a best model for checking the impact of time variant variables
on agricultural trade of Pakistan with United Arab Emirates. According to the
results of fixed effect model, GDP of Pakistan is positively affecting the trade
between Pakistan and its trading partners. The coefficient of GDP of
Pakistan is 0.75 (Table 3). One percent increase in GDP of Pakistan would
increase 0.75 percent agricultural trade between Pakistan and United Arab
Emirates. Which is highly significant. The value of estimated coefficient of
GDP of trading partner of Pakistan was 0.69 which means that one percent
increase in GDP of United Arab Emirates, would increase of 0.69 agricultural
trade of Pakistan with United Arab Emirates. According to Mohmand and
Wang (15) the coefficient of GDP in gravity model of trade was positive and
highly significant, which implies more trade of Pakistan with larger
economies. Khan and Saqib (13) concluded positive and significant

J. Agric. Res., 2016, 54(4)

794 I. Javed et al.

relationship between GDP and exports. Zarzoso (21) showed that income of
exporters and importers had a positive impact on bilateral trade. Roy and
Rayhan (17) carried out research on trade where all estimated coefficients on
the levels of GDPs were highly significant at 1 percent level with expected
sign. Their fixed effect model was significant and estimated that with one
percent increase of GDP, import demand of Bangladesh increased by 0.95
percent and export supply of country J increased by 1.61 percent. According
to Iqbal (11) GDP of Bangladesh was positively correlated with trade.
Population of Pakistan is negatively affecting the agricultural trade of
Pakistan with its trading partners. The estimated value of coefficient of
population of Pakistan is -1.24 by the fixed effect model. This effect is highly
significant and it means that one percent increase in the population of
Pakistan decreases 1.24 percent trade of Pakistan with United Arab Emirates
(Table 3). Zarzoso (21) found that population of export country had negative
impact on exports showing a positive absorption effect, while the population
of importer had a positive effect on exports. It indicated that bigger
economies import more than small economies (21).

The population of Pakistan carries the negative sign. It implies that

population variable has the trade inhibiting affect. It implies that a larger
population size may be treated as large resource endowments and thus an
indication of self-sufficiency and thereby less reliance on international trade.
Lubinga and Kiiza (14) conducted a research on agricultural trade flow and
observed a negative impact of population of trading partners on bilateral
trade flow.

The population of trading partners of Pakistan positively affected agricultural

trade of Pakistan. This impact is highly significant. It indicates that one
percent increase in the population of United Arab Emirates would increase
0.63 percent agricultural between Pakistan and United Arab Emirates. All
values of VIF were less than 10 which showed no multicollinearity existed in
the data set. (Table 4).
Table 4. Multicollinearity for the fixed effect model.

Variable VIF
DLGDPJ 8.57
DLGDPI 7.56
LPOPI 6.92
LPOPJ 1.22
Source: Author’s estimations.

J. Agric. Res., 2016, 54(4)

 Agricultural trade between Pakistan and UAE by using gravity model 795

To solve the problem of heteroscedastisity, HAC standard errors were used

during the estimation of results. Autocorrelation in the model is analyzed by
the value of Durban Watson test for the estimated model. The value close to
2 is the sign of no autocorrelation in the model. If the value be less than two
means positive autocorrelation and the value more than 2 indicates the
negative autocorrelation in the model. So by the value given in model of fixed
effect showed that there was no problem of autocorrelation in the estimated
model.

After estimating the F test and Housman specification test both FEM and
REM were estimated. F test and Hausman specification tests were in favor of
FEM so this model was considered best for the time variant variables. For the
time invariant variables, REM was considered best, by Breusch-Pagan LM
test to analyze the impact of all variables under consideration on agricultural
trade between Pakistan and United Arab Emirates.

Estimation of Gravity Model by Random Effects Model: The random

effects treatment does allow the model to contain observed time invariant
characteristics (8). Roy and Rayhan (17) used the random effect model to
estimate the impact of time invariant variables in the panel data analysis by
using gravity model. Multicolinearity in the random effect model is shown by
the value of VIF (Table 5).

Table 5. Multicollinearity for the model of random effects.

Variable VIF
DLGDPJ 6.13
DLGDPI 7.50
LPOPI 6.32
LPOPJ 3.21
CULTURE 3.65
BORDER 3.23
LDISIJ 5.31
Source: Author’s estimations

All variables included in the model of random effects have no problem of

multicolinearity because the values of variance inflation factor is less than
ten. The value of 1.97 of Durban Watson test in the model of random effects
showed that there was no problem of autocorrelation in the estimated model.

GDP of Pakistan is significantly affecting agricultural trade between Pakistan

and its trading partners. It means that one percent increase in GDP of

J. Agric. Res., 2016, 54(4)

796 I. Javed et al.

Pakistan would increase about one percent agricultural trade with United
Arab Emirates (Table 6). GDP of trading partners of Pakistan significantly
affected agricultural trade of Pakistan with its trading partners. One percent
increase in GDP of United Arab Emirates would increase 0.51 percent
agricultural trade with Pakistan.

Table 6. Estimated gravity model by random effects model.

Variable Coefficient Std. Error t-Statistic Prob.

*
C 5.53 2.73 2.02 0.043
***
LGDPJ(-1) 0.51 0.06 8.42 0.000
***
LGDPI(-1) 1.00 0.28 3.56 0.000
NS
LPOPI -1.13 0.73 -1.56 0.119
***
LPOPJ 0.23 0.07 3.28 0.001
***
CULTURE 1.27 0.18 6.89 0.000
***
BORDER -2.34 0.23 -10.06 0.000
NS
LDISIJ -0.18 0.17 -1.02 0.306
Adjusted R-squared 0.83
Durbin-Watson stat 1.97
F-statistic 260.79
Prob(F-statistic) 0.000
HAC standard errors are used for removal of heteroscedastisity, *Significant at 5% level of
confidence, ***Highly significance, NS = Non Significant

Population of Pakistan has a negative effect on agricultural trade of Pakistan

that is not significant. According to estimated results of random effects, one
percent increase in population of Pakistan would decrease 1.13 percent
agricultural trade between Pakistan and United Arab Emirates. Howerver,
population of major trading partner of Pakistan has a positive effect on the
agricultural trade. This impact was highly significant which described that one
percent increase in the population of United Arab Emirates would increase
0.23 percent agricultural trade between Pakistan and United Arab Emirates.

According to the estimation of random effect model, distance was negatively

affecting agricultural trade. It showed that, an increase in distance between
trading partners has a negative effect on agricultural trade but this impact
was not significant. Rahman (16) found that transportation cost had a
significant impact on the Bangladesh’s trade and that effect was negative as
expected. Roy and Rayhan (17) analyzed that estimated coefficient on
distance was highly significant at 1 percent level with the expected sign and
showed estimated value -0.61 percent as a result of 1 percent increase in
bilateral distance between these two countries. Cultural similarities between
trading partners has a positive impact on the agricultural trade and this was

J. Agric. Res., 2016, 54(4)

 Agricultural trade between Pakistan and UAE by using gravity model 797

highly significant relationship. The similar culture of Pakistan with its major
trading partners have 1.27 percent more trade as compared to the trading
partners which have no cultural similarities with Pakistan (Table 6).

Joint border is assumed to have a positive impact on the bilateral trade but in
case of Pakistani agricultural trade the fact is that, Pakistan has not good
relations with the countries that have joint border, so this impact was negative
and highly significant. The joint border countries of Pakistan have 2.3 percent
less agricultural trade as compared to the countries which have no joint
border. Mohmand and Wang (15) described that trade environment of
Pakistan was facing chronic trade deficit by using the gravity model.
According to them results non-significance value of common border variable
was showing less trade between the nations of close proximity to Pakistan,
like India, Iran, and Bangladesh. Pakistan needs to build good relationship
with its neighbors to gain maximum benefits by trade being a neighbor.

Considering time invariant variables there may be two models for the
analysis. One is the random effects and the other is Pooled OLS model.
According to Breusch and Pagan LM test of model specification, random
effect model was selected as a best model against the Pooled OLS model for
the analysis of bilateral trade between Pakistan and United Arab Emirates.

CONCLUSION

GDP of Pakistan has a positive and significant impact on agricultural trade

with United Arab Emirates. Pakistan needs to find large economies for the
trade of agricultural products. GDP of United Arab Emirates has also a
positive and significant impact on agricultural trade between Pakistan and
United Arab Emirates. Population of United Arab Emirates has positive
impact on agricultural trade. It is recommended that Pakistan should try to
focus more on the economies having large population as a trading partner for
agricultural products. Negative impact of increasing Population of Pakistan
on agricultural trade suggests that Pakistan should try to improve the labour
qualities and skills that may be favorable in agricultural trade growth as well
as development of economy. The negative impact of distance and joint
border between trading partners on agricultural trade suggests that Pakistan
should try to improve the trade environment with the countries that are close
to Pakistan and also with its neighbours. After positive and significant impact
of cultural similarities on agricultural trade, it is recommended that Pakistan
should try to concentrate on those countries having culture similar to
Pakistan.

J. Agric. Res., 2016, 54(4)

798 I. Javed et al.

REFERENCES

1. Anon. 2012. Economic Survey of Pakistan (2011-12). Economic

Advisor’s Wing, Finance Division, Government of Pakistan, Islamabad
Pakistan.
2. Anon. 2013. Economic Survey of Pakistan (2012-13). Economic
Advisor’s Wing, Finance Division, Government of Pakistan, Islamabad
Pakistan.
3. Anon. 2013. Trade Maps: Trade Statistics for international business
development. International trade centre, Geneva Switzerland.
4. Antonucci, D. and S. Manzocchi. 2006. Does Turkey have a special
trade relation with the EU: A Gravity Model Approach,” Economic
Systems. 30(2): 157-169.
5. Baltagi, B. H. 2005. Econometric Analysis of Panel Data. 3rd Ed.
Cambridge University Press, New York.
6. Breusch, T. S. and A.R. Pagan. 1980. The lagrange multiplier test and
its applications to model specification in econometrics. Review of
Economic Studies. 47(1):239-253.
7. Egger, P. 2002. An econometric view on estimation of gravity models
and the calculation of trade potentials. World Economy. 25 (2):297-312.
8. Granger, C.W.J. and P. Newbold. 1974. Spurious regressions in
econometrics. J. Econ. 2:111-120.
9. Greene, W. 1997. Econometric Analysis 5th Edition. Upper Saddle
River, N.J: Prentice-Hall.
10. Hausman, J. A. 1978. Specification tests in econometrics,
Econometrica. 46:1251-1272.
11. Iqbal, M. H. 2014. Determinants of bilateral trade between Bangladesh
and the European Union: Approach of gravity model under the panel
data. Issues in Business Management and Economics.2 (5):087-093.
12. Kepaptsoglou, K., M. G. Karlaftis and D. Tsamboulas. 2010. The gravity
model specification for modeling international trade flows and free trade
agreements effects. The Open Economics Journal. 3:1-13.
13. Khan, A. and N. Saqib. 1993. ‘Exports and Economic Growth: The
Pakistan experience’. Int. Econ. J. 7(3):53-64.
14. M. H. Lubinga and B. Kiiza. 2013. Exchange rate uncertainty and
bilateral trade flows: Insights from Uganda’. Business and Economic
Research. ISSN 2162-4860 2013, Vol. 3, No. 1 www. macrothink.
org/ber 227.
15. Mohmand, Y. T. and A. Wang. 2013. The gravity of Pakistan’s export
performance in stratified sampling. Pak. J. Statist. 29(2):203-216.

J. Agric. Res., 2016, 54(4)

 Agricultural trade between Pakistan and UAE by using gravity model 799

16. Rahman, M. 2003. ‘A Panel Data Analysis of Bangladesh’s Trade: The

Gravity Model Approach’. Available at: http: // www. etsg. Org /
ETSG2003 / papers / rahman. pdf, accessed on March 26, 2011.
17. Roy, M. and M. I. Rayhan. 2012. Import Flows of Bangladesh: Gravity
model approach under panel data methodology’. Dhaka Univ. J. Sci.
60(2):153-157.
18. Suvankulov, F. and W. Ali. 2012. Recent trends and prospects of
bilateral trade between Pakistan and Turkey: A gravity model approach.
J. Int. Global Economic Studies. 5(1): 5772.
19. Tinbergen, J. 1962. Shaping the World Economy: Suggestions for an
International Economic Policy. The Twentieth Century Fund. World
Trade Organization. New York. Website: www.wto.org
20. Wooldridge, J. M. 2002. Econometrics Analysis of Cross Section and
Panel Data. The Press Cambridge, Massachusetts, London. p. 288-
291
21. Zarzoso, I. M. 2003. ‘Augmented gravity model: an empirical application
to Mercosur-European Union Trade Flows’. J. Appl. Econ. 6(2):291-316.

Received: October 21, 2014 Accepted: August 11, 2015

***

CONTRIBUTION OF AUTHORS:

Iqbal Javed : Conducted the research, analyzed data and prepared

writeup.
Muhammad Ashfaq : Supervisor
Sultan Ali Adil : Co-supervisor
Khuda Bakhsh : Co-supervisor

J. Agric. Res., 2016, 54(4)

View publication stats