CC BY
13COINTEGRATION ANALYSIS OF ECONOMIC GROWTH PARAMETERS BETWEEN RUSSIA
AND AZERBAIJAN
Huseynova S.
Ph.D., Associate Professor of Mathematical Economics
Baku State University DOI: 10.5281/zenodo.7618240
ABSTRACT
The purpose of the analyze the integration process economic growth parameters between Russia and Azerbaijan. Based on the 28-year statistical indicators of economic growth parameters, an econometric study of dependence was conducted in the time period covering the years 1994-2021. In research is applied the Johansen cointegration test, vector-error-correlation-model (VECM) and variance-decomposition (VDC).Vector-error-correlation-model (VECM) adequate is tested cert. The result was further substantiated by the tests based on Johansen cointegration and VECM procedures, showing significant long-run economic relations. The research applies the cointegration techniques in the context of CIS republics.
Keywords: Economic growth, Azerbaijan, Russia, Johansen cointegration test, VECM, impulse response function, variance decomposition.
Introduction. Although regional integration is of great importance in modern times, from the perspective of individual states, the main factor for this integration to be successful is that it first of all meets the economic interests of the states. On the other hand, how successful the replication of integration processes in other regions will be in any region depends on the realities and conditions existing in and around that region. For the success of the energy transaction process in the region, the strategy in the South Caucasus has an important impact on the development of resources, transit opportunities and economic potential.
Recent Publications. In this work, we will analyze the issue of cointegration of Azerbaijan-Russia trade relations. In this aspect, research works [2-9] can be mentioned in the problem statements. The purpose of our research is stated in [9] that for the same statistical data, the selection of a new specification of the model to increase the accuracy of the forecast compared to [9] is to justify that the errors of the statistical values are smaller than the error of the model in [9].In recent years, it can be seen that more researchs of the econometric analyses about interraletions of growth parametres between the countries[4,5]. In the conducted researches, a model was established by evaluating the GDP of country on the components of the trade turnover of the countries, analyzing the theoretical-methodological bases of macro variables [1-10]. In research paper is used annual data from State Statistics Committee of Azerbaijan and World Bank [11,12] . Eviews 8 software package was used for calculating econometrical procudures
Methodology. The aim of this article by the econometric methods to analyses the interrelation of growth parametres of Azerbaijan , Russia during the period 1994-2021. In this article, we study the growth indicators by the natural logarithmic values. All time series will be transformed into natural logarithmic ones. This
transformation makes it possible to more visually represent the relationship between the expected results. The first differences of the natural logarithms are an approximation of the similarity growth rates.
The natural logarithm of the Russia trade turnover between Azerbaijan (LNRUST) is dependent from natural logarithm of the GDP of Russia (LNGDPRUS) and natural logarithm of the GDP of Azerbaijan (LNGDPAZ). The form of the multivariate linear regression model will be as follows:
LN_RUSSIA_T = 0.492996856788*LN_GDP_AZERBAIJAN + 0.441464340108*LN_GDP_RUSSIA + 0.0172566963155*LN_RESID - 3.45375419309 (1)
Upon applying the Augmented Dickey-Fuller (ADF) tests for stationarity (unit root), [9] shows the results of the test for all variables both in levels and second differenced forms. As can be seen, all test results suggest that variables were second differenced stationary, we then proceeded to perform a cointegration analysis having satisfied this condition.
The verification of the causal relationships between the factors for lag values m = 1, 2, 3, 4 was carried out through the Granger Test. The Granger Causality Test, with the exception of one directions, confirmed the presence of a two-way causal relationship, which indicates the existence of a third variable, which is the real cause of the change in the two considered variables. Only for lag m = 4, the causal relationships between ALNGDPAZ and ALNRUT is opposite. Here, A denotes the difference operator of the corresponding variable.
To establish the cointegration relationship between variables, we constructed an unrestricted vector autoregression (VAR) model for LNRUT, LNGDPAZ and LNGDPRU. The unrestricted co-integration rank test of Max-Eigenvalue as shown in Table 2 as follows:
Table 2.
__Result of Max-Eigenvalue test__
Hypothesys Alternative hypothesis Statistical Max-Eigenvalue Critical value 1% Prob.
Ho r=0* Ha r >0 46.28445 30.83396 0.0000
Ho r=1 Ha r >1 21.14861 23.97534 0.0275
Ho r=2 Ha r >2 1.818572 16.55386 0.9786
The unrestricted co-integration rank test of trace as shown in Table 2 as follows:
Table 3.
Result of Trace test
Hypothesis Alternative hypothesis Trace statistics Critical value 1% Prob
Ho r=0* Ha r >0 69.25163 49.36275 0.0000
Ho r=1 Ha r >1 22.96718 31.15385 0.1103
Ho r=2 Ha r >2 1.818572 16.55386 0.9786
Vector error correction model (VECM) with lag shocks and the rate of its recovery. Performing the pro-equal to 2 and rank 1, which expresses the long-term cedures of the Eviews 8 program, the following error equilibrium relationship of variables and the authentic- correction equation was found for the second-order dif-ity of their correlation, which makes it possible to ferences of the logarithmic values of Azerbaijan's GDP: measure deviations from equilibrium in the event of
D(DLNRUST) = - 0.290077133661 *(DLNRUST(-1) + 2.78881772296*DLNGDPRU(-1) -1.85742294647*DLNGDPAZ(-1) - 0.461510661477*DLNRESID(-1) + 0.0312610273533 ) -0.732694182304*D(DLNRUST(-1)) - 0.327588146947*D(DLNRUST(-2)) -0.168290149682*D(DLNGDPRU(-1)) + 0.139967205345*D(DLNGDPRU(-2)) + 0.281841167543*D(DLNGDPAZ(-1)) - 0.546582599288*D(DLNGDPAZ(-2)) -0.115313674768*D(DLNRESID(-1)) - 0.0765403164374*D(DLNRESID(-2)) + 0.0148990588517
D(DLNGDPRU) = - 0.392267957358*(DLNRUST(-1) + 2.78881772296*DLNGDPRU(-1) -1.85742294647*DLNGDPAZ(-1) - 0.461510661477*DLNRESID(-1) + 0.0312610273533 ) + 0.241378953426*D(DLNRUST(-1)) + 0.185119549514*D(DLNRUST(-2)) + 0.141800514658*D(DLNGDPRU(-1)) - 0.160655828857*D(DLNGDPRU(-2)) -0.613526363404*D(DLNGDPAZ(-1)) - 0.333968836648*D(DLNGDPAZ(-2)) -0.13359571784*D(DLNRESID(-1)) - 0.027770586531*D(DLNRESID(-2)) + 0.0392988670524
D(DLNGDPAZ) = - 0.0366296719825*(DLNRUST(-1) + 2.78881772296*DLNGDPRU(-1) -1.85742294647*DLNGDPAZ(-1) - 0.461510661477*DLNRESID(-1) + 0.0312610273533 ) -0.0204215036463*D(DLNRUST(-1)) + 0.0480200599328*D(DLNRUST(-2)) + 0.0698119947368*D(DLNGDPRU(-1)) + 0.0903524564785*D(DLNGDPRU(-2)) -0.517420248631*D(DLNGDPAZ(-1)) - 0.678454182714*D(DLNGDPAZ(-2)) -0.00517585437399*D(DLNRESID(-1)) + 0.0237083737271*D(DLNRESID(-2)) + 0.0188605794964
The adequacy of VECM is tested Var residual serial correlation LM test, Var residual normality test, white test for heteroscedasticity:No cross terms Table 4,5,6 is shown tests result.
Table 4.
Result of Var residual serial correlation LM test
Null Hypothesis: no consistent correlation with lag order h
Date: 01/27/23 Time: 17:40
Sample: 1994 2021
Included observations: 24
Lags LM-Stat Prob
1 16.05320 0.4493
2 25.57575 0.0603
Table 5.
Result of VAR Residual Normality Tests VAR Residual Normality Tests_
Orthogonalization: Cholesky (Lutkepohl)
Null Hypothesis: residuals are multivariate normal
Date: 01/27/23 Time: 17:47
Sample: 1994 2021
Included observations: 24
Component Skewness Chi-sq Df Prob.
1 0.244758 0.239626 1 0.6245
2 -0.608881 1.482942 1 0.2233
3 0.642626 1.651873 1 0.1987
Joint 3.374441 3 0.3374
Component Kurtosis Chi-sq df Prob.
1 2.457820 0.293959 1 0.5877
2 2.893919 0.011253 1 0.9155
3 3.266936 0.071255 1 0.7895
Joint 0.376467 3 0.9451
Component Jarque-Bera df Prob.
1 0.533585 2 0.7658
2 1.494195 2 0.4737
3 1.723128 2 0.4225
Joint 3.750908 6 0.7103
White test for heteroscedasticity:No cross terms
Table 6.
_ Result of White test for heteroscedasticity:No cross terms__
Dependent R-squared F(12,11) Prob. Chi-sq(12) Prob.
res1*res1 0.779897 3.248046 0.0302 18.71752 0.0956
res2*res2 0.315175 0.421874 0.9232 7.564191 0.8182
res3*res3 0.364328 0.525377 0.8580 8.743878 0.7246
res2*res1 0.640229 1.631248 0.2131 15.36549 0.2221
res3*res1 0.407638 0.630812 0.7798 9.783323 0.6350
res3*res2 0.247611 0.301675 0.9748 5.942668 0.9189
Analyse of Impulse-response functions and forecast Error Variance Decomposition (FEVD). Impulse-response functions track the dynamic effect on a system "shock" or change in input. Although impulse response functions are used in many fields, they are particularly useful in macroeconomics for a number of reasons. How the results change in the face of exogenous variables of the established model can be shown through the impulse response function. It can be used to predict the consequences of economic changes in the macroeco-
nomic framework. It can be used to model the economic integration process. Enabling structural constraints are used.
We expect stationary time series to show that shocks to the system are persistent and that the system converges over time. When the system converges, it may or may not converge to the initial state, depending on the constraints imposed on our structural VAR model. The results of testing on 10-year time horizons are described in Figure 2:
Response of DLNRUST to Cholesky One S.D. Innovations
DLNRUST - DLNGDPAZ - DLNGDPRU |
Response of DLNGDPAZ to Cholesky One S.D. Innovations
DLNRUST - DLNGDPAZ - DLNGDPRU |
Response of DLNGDPRU to Cholesky One S.D. Innovations
— DLNRUST - DLNGDPAZ - DLNGDPRU |
Figure 1. Result of impulse response function
As a result, when a positive aggregate supply shock occurs, output approaches a higher level than before the shock. The resulting impulse response functions have a clear modeling procedure: Estimate a structural VAR model. Predict the impulse response functions along with their confidence intervals for a given time horizon.
The Forecast Error Variance Decomposition (FEVD) show you how much of the future uncertainty
of one time series is due to future shocks into the other time series in the system. We could proof that if all confidence interval of impulse response function graphs usually contains the zero horizontal axis, which means that the response is insignificant at 95% confidence level and therefore in Variance decomposition table, the variance in the forecast error of all other variables is not completely explained by the variable alone.
8060-
Variance Decomposition of DLNRUST
123456789 10
- DLNRUST -DLNGDPRU -DLNGDPAZ]
Variance Decomposition of DLNGDPRU
DLNRUST -DLNGDPRU -DLNGDPAZ
Variance Decomposition of DLNRUST
7060 50 40 302010 0
Figure 2. Result of variance decomposition function
In result variance decomposition is used more norrowly for a specific tool for interpreting the relations between LNRUT and LNGDPAZ, LNGDPRU variables decribed by vector autoregressive (VAR) models.
Result:
Today we've provided an intuitive look at impulse response functions and variance decompositions. These two multivariate time series tools are fundamental applications of the structural VAR model. In result of impulse response functions describe the reaction of endogenous economic growth variables at the time of the shock and over subsequent points in time. Variance decomposition analysis allows in research partitioning the total variance in an outcome variables. This paper has investigated of Johansen's maximum eigenvalue and trace tests for cointegration.
Established VECM and checked adequate by the
tests.
As a result of the conducted researches, effective regulation of trade turover of Azerbaijan - Russia is impacted to GDP of Azerbaijan and GDP of Russia .By the econometric analyses we show that development of econometrically based recommendations that allow to conduct dynamic analyzes for the long-term sustainable development of the country's economy are considered as one of the most urgent and priority issues of our time.
Azerbaijan and Russia economic relationship effective result is impacted to development growth of
country .Effective trade turnover between the countries gives opportinity to Russia is taking advantage of investment opportunities in Azerbaijan, and this is also important for Azerbaijan as a foreign investment.The relationship is based on political dialogue, reliable partnership, alliance, which acts as a solid pillar for all other cooperative relations.These relations strengthen the interests and trends of stability, security and sustainable development in the region.
References
1. Ali, Ifzal, and Hyun Hwa Son. 2007. "Measuring Inclusive Growth." Asian Development Review, Asian Development Review 24 (1): 11-31.
2. Orudzhev, E., & Alizade, A. (2021). Cointegra-tion analysis of the impact of Azerbaijan and Ukraine GDPs on the trade turnover between these countries. Journal of International Studies, 14(3).
3. Orudzhev, E. G., & Huseynova, S. M. About One Cointegration Issue Of Trade Relations Azerbaijan, Russia, Belarus. Journal of Advanced Research in Dynamical and Control Systems (JARDCS), 12(6), 1385-1394.
4. Orudzhev, E. G., & Huseynova, S. M. (2020). On one co-integration issue of trade links of Azerbaijan, Russia, Belarus and Kazakhstan. Statistics and Economics, 17(2), 29-39.
5. Huseynova, S. (2020). THE ECONOMETRIC ANALYSIS OF THE INTERRELATION BETWEEN
40
20
0
80
60
40
20
0
2
3
4
5
6
7
8
9
0
80
4
0
THE GDP OF AZERBAIJAN, RUSSIA, BELARUS AND KAZAKHSTAN. InterConf.
6. Huseynova, S., & Babazade, G. (2022). COiNTEGRATiON ANALYSiS OF iMPORTEXPORT OPERATiONS OF THE REPUBLiC OF AZERBAiJAN iN RELATiON TO THE EXCHANGE RATE OF THE MANAT AND THE iNCOME OF THE POPULATiON. InterConf,(105), 45-54.
7. Huseynova, S. (2022). THE ECONOMETRIC ANALYSIS OF THE INTERRELATION BETWEEN THE INCLUSIVE INDICATORS OF AZERBAIJAN, RUSSIA AND KAZAKHSTAN. TURAN-SAM, 14(56), 118-124.
8. HUSEYNOVA, S. M., & BABAZADE, G. S. ECONOMETRIC ANALYSiS OF THE ELASTiCiTY OF iMPORT-EXPORT OPERATiONS OF THE REPUBLiC OF AZERBAiJAN iN RELATiON TO THE EXCHANGE RATE OF THE MANAT AND
THE iNCOME OF THE POPULATiON. EURASIAN UNION OF SCIENTISTS. SERIES: ECONOMIC AND LEGAL SCIENCES Учредители: ООО" Ло-гика+", (2), 3-8.
9. S. Huseynova Econometric analysis of economic growth parameters between Russia and Azerbaijan, XVIII. International Balkan and Near Eastern Congress Series on Economics, Business and Management-Ohrid/Republic of North Macedonia, Proceedings of University of Agribusiness and Rural Development/Bulgaria University "St. Kliment Ohridski", Faculty of Economics - Prilep/Republic of North Macedonia p.80-86,26-27 november,2022
10. Krugman, Paul, and Richard E. Baldwin, "The Persistence of the U.S. Trade Deficit Brooking Paper on Economic Activity I (1987), pp. 1-5
11. www.worldbank.com
12. www.stat.gov.az
ДЕЯК1 ПИТАННЯ УПРАВЛ1ННЯ СИСТЕМОЮ ПУБЛ1ЧНИХ ЗАКУП1ВЕЛЬ В еВРОПЕЙСЬКОМУ СОЮЗ1 ТА В УКРА1Н1
Левон С.Б.
Головний спецгалгст в1ддту мотторингу та контролю якостг аудитгв Департаменту стратеггч-
ного розвитку та методологИ Рахунковоi палати
SOME ISSUES OF MANAGING THE PUBLIC PROCUREMENT SYSTEM IN THE EUROPEAN
UNION AND IN UKRAINE
Levon S.
Chief Specialist, Audit Quality Monitoring and Control Unit, Strategic Development and Methodology Department, Accounting Chamber
DOI: 10.5281/zenodo.7618242
АНОТАЦ1Я
У статп визначено прюритетш напрями розвитку системи управлшня публiчними закутвлями в Укрш'ш та здшснено аналiз досвщу кра!н £вропейського Союзу. Охарактеризовано базовi положення стан-дартизованих нормативних акпв, яш регулюють закутвельний процес в кранах £вропейського Союзу, а саме: Типового закону ЮНС1ТРАЛ, Угоди про СОТ та Директив £С. Проаналiзовано Директиви £вропей-ського Союзу 2014/24/GC та 2014/25/GC, якi е першорядними законодавчими актами у регулюваннi публiчними закупiвлями в £вропейському Союзi.
Установлено, що законодавство Украши та £вропейського Союзу у сферi публiчних закутвель мають спiльну мету - проведения прозорих тендерiв та забезпечення економiчноi ефективностi публiчних за-купiвель. З'ясовано, що законодавство £вропейського Союзу покликане регулювати набагато складнiшi вiдносини, шж регулюе законодавство Украiни на тепершнш час.
Уточнено прiоритетнi напрями розвитку правового забезпечення функцюнування нацiональноi системи управлшня публiчними закупiвлями. Обгрунтовано доцiльнiсть замiни в чинному законодавствi Украiни про публiчнi закупiвлi поняття "замовника", оновлення процедури конкурентного дiалогу та пе-реговорiв з кшькома учасниками, поширення практики щодо ведення електронних каталогiв, запро-вадження нових правил щодо заповнення тендерно1' документаци та тендерних процедур i оголошень, унормування правил допорогових закупiвель. Аргументовано доцiльнiсть закрiплення на законодавчому рiвнi нового принципу пропорцiйностi закупiвель за бюджетш кошти з урахуванням ст. 36 Директиви 2014/25/6С.
ABSTRACT
The article identifies the priority directions for the development of the public procurement management system in Ukraine and analyzes the experience of the European Union. It outlines the basic provisions of the standardized regulations that govern the procurement process in the European Union, namely: the UNCITRAL Model Law, the WTO Agreement, and EU Directives. European Union Directives 2014/24/EU and 2014/25/EU, which are the primary legislative acts to regulate public procurements in the European Union, have been analyzed.
It has been identified that the public procurement laws of Ukraine and the European Union have a common goal - transparent tenders and economic efficiency of public procurements. The legislation of the European Union
