Theoretical and practical aspects of the development of e-commerce in the innovative development of the economy have been studied by many foreign scholars, including P.F.Druker [3], B.Twiss [4], Y.Shumpeter [5], R.Fostr, Webster Frank [6], A. Dinis, Y Hsu, K.Oppenlender [7], Y.Hsu [8] et al.
Among the scientists of the Commonwealth of Independent States: P.N. Zavlin [9], L.P. Goncharenko [10], L.M. Goxberg [11], A.K. Kazantsev [12], B.Z. Milner, Yu. Studied in the scientific works of V. Yakovets.
Methodical aspects of establishing a control system over compliance with principles of decent work and social security in textile enterprises were researched by G.K. Abdurakhmanova and others [16; 18] Innovative development of Uzbekistan agroindustrial complex were dedicated works of Yldashev, N., Nabokov, V. I., Nekrasov, K. V. [19;20]. Estimation methodology of efficiency of production capacity management at textile enterprises were investigated by Kirill K. and others [21]. Role of Managing Industrial Stocks in Increasing of Textile Enterprises Capacity were studied by Tursunov B. and others [1; 17; 22].
Theoretical and practical problems of effective use of information and communication technologies from Uzbek scientists B.B. Abdullaev [13], I. Abduraimov, S.S. Gulamov [14], B.B. Berkinov, A.N. Aripov [15] The multi-factor econometric model based on e-commerce indicators, studied in the scientific works of T.Z. Teshabaev, Sh.A. Tursunov, R.I. Nurimbetov and others, is tested by a number of criteria. Based on the tested model, it will be possible to forecast the performance of the e-commerce sector in the coming periods.
Analysis and results
The e-commerce sector in Uzbekistan is developing. Today, the volume of e- commerce is about one percent of the country's GDP. In order to further develop e- commerce, it is necessary to identify the factors that affect it, to study the appearance of links between them and on this basis to forecast future periods.
A number of factors affect the e-commerce sector in Uzbekistan. These include the number of Internet users, the cost of Internet service tariffs, the number of online stores, e-commerce transactions, the volume of e-commerce transactions, total transactions through POS terminals, the number of plastic cards, the volume of plastic card transactions, the number of ATMs and kiosks, etc. a number of factors can be cited.
The following factors were selected to create a multifactor econometric model for the activities of the e-commerce sector (semi-annual data for 2010-2019): the result - the volume of e-commerce services, bln. soums - (Y), influencing factors - number of Internet users, mln. person - (X1), cost of Internet services tariffs, USD, - (X2), number of online stores, unit - (X3), e-commerce transactions, mln. unit - (X4), total transactions through POS terminals, mln. unit - (X5), number of plastic cards, mln. units - (X6) and the number of ATMs and kiosks, units - (X7).
Since the unit of measurement of the variables is different, and to better explain the multi-factor econometric model interpretation, we logarify all the factors.
Descriptive statistics on factors were first conducted when constructing a multifactor econometric model. To do this, a special econometric modeling program - Eviews 10 was used. The results of the descriptive statistics are presented in Table 1 below.
Factor analysis` statistics
From the data in the table you can see the average value (mean), median (maximum), maximum and minimum values (maximum, minimum) of each factor. In addition, the standard deviation of each factor (std. Dev. (Standard Devation) - the standard deviation coefficient indicates how much each variable deviates from the mean) is given.
Because the asymmetry coefficients of some factors (lnX1, lnX2, lnX6, and lnX7) were negative, the left tail of their graphs was pushed to the left of the theoretically normal distribution graphs. Since the asymmetry coefficients of the other factors were positive, their right tail was pushed to the right.
A correlation analysis should be performed to select factors for a multifactor econometric model. Among the factors for this are the private and double correlation coefficients. The matrix of specific and double correlation coefficients between factors is given in Table 2 below.
Correlation matrix between fators Covariance Analysis: Ordinary
t-Statistic Probability
As can be seen from this table, the specific correlation coefficients are the density of the relationship between the resulting factor and the factors influencing it. Hence, the specific correlation coefficients indicate that there is a strong correlation between the resulting factor (volume of e-commerce services, lnY) and the influencing factors, i.e. the value of the specific correlation coefficients is greater than 0.7.
In addition, Table 2 also contains double correlation coefficients, which show the bond densities between the influencing factors (lnXi, lnXj). The most important thing here is that the influencing factors should not be closely related to each other. That is, there should be no multicollinearity. If the value of the double correlation coefficient between the two influencing factors is less than 0.7, multicollenity is said to be non-existent. From the data in Table 2, it can be seen that the bond densities between the influencing factors are not greater than 0.7. Hence, there is no multicollenearity among the influencing factors. Another way to check for the absence of multicollenearity between influencing factors is to calculate VIF (Variance Inflation Factors) coefficients.
If there is multicollarity between influencing factors, then VIF> 10. According to the results, the VIF coefficients of all influencing factors are less than 10. Hence, this also indicates that there is no multicollinearity between influencing factors, such as correlation analysis.
Using the data in Table 2, we present a mathematical view of the multifactor econometric model:
lny6,87640,2978lnx1 0,0529lnx2 0,1726lnx3 0,4794lnx4 (1)
0,1092lnx5 0,4851lnx6 0,1999lnx7
The calculated multi-factor econometric model shows that if the number of Internet users (lnx1) increases by an average of 1.0%, the volume of e-commerce services (lny) can increase by an average of 0.2978% (as a result of online shopping from online stores). effective). In Uzbekistan, the increase in tariffs for Internet services (lnx2) increased by an average of 1.0%, while the volume of e-commerce services (lny) decreased by an average of 0.0529%. (This inverse relationship is also reflected in the correlation matrix between the factors) (decrease in the cost of internet service tariffs will allow providers to provide more types of services to both e-commerce and Internet users). In our country, the number of online stores (lnx3) increased by an average of 1.0%, the volume of e-commerce services (lny) increased by an average of 0.1726% (as a result of the increase in online stores, citizens can buy home appliances, books, office equipment, construction materials, etc.). purchasing power will increase). An increase of e-commerce transactions (lnx4) by an average of 1.0% can lead to an increase in the volume of e-commerce services (lny) by an average of 0.4794% (in which case the transactions can be distributed among a number of services). An increase in total transactions (lnx5) through POS terminals by an average of 1.0% could lead to an increase in the volume of e- commerce services (lny) by an average of 0.1092% (not only trades but also cash withdrawals). Among the population, plastic cards can lead to an average increase in the number of corporate plastic cards in enterprises (lnx6) by an average of 1.0 percent, the volume of e-commerce services (lny) by an average of 0.4851 percent (electronic payment systems by citizens or businesses plastic cards Click, Payme, Performs e-purchases through Unipay and other systems). An increase in the number of ATMs and kiosks (lnx7) in Uzbekistan by an average of 1.0% may lead to an increase in the volume of e-commerce services (lny) by an average of 0.1999%.
To check the quality of the multifactor econometric model (3.4), we check the determination coefficient. The coefficient of determination indicates the percentage of the factor that is included in the model. The calculated determination coefficient (R2 - R-squared) is 0.9844. This shows that 98.44% (3.4) of the volume of e- commerce services are factors included in the multifactor econometric model. The remaining 1.56 percent (1.0-0.9844) is due to factors not taken into account.
|
