Econometric modeling of the calculation of gross domestic product with the expense method

. The results of the correlation between the factors affecting the gross domestic product were calculated on the basis of the methodology of the national accounting system, a regression equation was identified, and short-term alternative forecasting issues were clarified using a statistical index of GDP growth in 1997-2019 of the Republic of Uzbekistan in this article.


Introduction
It plays an important role in ensuring the stability of the country's economy, statistical assessment of development factors and trends at different levels, development of a statistical index system, development of statistical models of economic development strategies and improving the scientific and methodological basis of statistical forecasting. To conform to the activities of the State Statistics System directly related to the third and fourth of these areas in line with modern requirements and international standards, the President of the Republic of Uzbekistan adopted Resolution PQ-№3165 of July 31, 2017, "On measures to improve the State Statistics Committee". The resolution sets tasks to radically improve the system of state statistics, to introduce modern methods of statistical analysis, indexes, evaluation criteria and reporting forms, tested in practice, widely used in international statistics, in accordance with international requirements and standards.
Furthermore, in his Address to the Oliy Majlis on the most important priorities for 2019, the President said that "To objectively assess the gross domestic product, we must fully implement the system of national accounts of the United Nations and the International Monetary Fund in the country from January 1, 2020" [1]. Effective implementation of these tasks requires the full development of the National Accounting System (NAS) in accordance with the standards adopted in international practice in Uzbekistan and the use of new methods of statistical data analysis for our national statistics in the introduction of SMEs in statistical practice.

Research literature
Theoretical-methodological issues on the methods of statistical accounting of macroeconomic indexes of the NAS have been deeply studied in the scientific works of foreign scientists R. Stoun, J. Keyns, V. Leontev, J. Marshall, and others [2][3][4][5].
The issues of indicating theoretical principles of NAS and international compari-sons of macroeconomic indexes were widely studied in scientific works of B.I. Bashkatov, V.I. Jerebin, V.K. Zaytsev and others, scientists of countries of the Com-monwealth of Independent States (CIS) [6][7][8].
Issues regarding the introduction of the national accounting system in Uzbekistan and the counting, analysis and forecasting of its macroeconomic indexes have been studied by scientific works scientists such as S.S. Gulomov, B.K. Goyibnazarov, B.J. Mirzanov [9][10][11][12].
However, in the scientific studies of the scientists mentioned above, the issues on the improvement of macroeconomic indexes of statistical accounts have not been widely studied in the new standard of the national accounting system in 2008. In this regard, based on the standard "NAS-2008", the statistical calculation of macroeconomy indexes of the country, the analysis of the factors influencing it, and the statistical forecasting of future development trends are important.

Research methodology
In the process of the research, the methods of empirical research, logistic function and correlative-regressive analysis were used.

Results and analysis
To quantify the impact of factors of production on the results of economic activity, the study of interrelated economic indexes of the NAS is of paramount importance. In this case, the indexes of gross production, gross domestic product and national income are selected as the result. It should be noted that the concept of "fac-tor and result" is comparatively accounted for; that is, based on the analysis of issues, these indexes are expressed as a factor and an outcome indicator.
The use of correlation-regression analysis to statistically study the interrelationships between the generalized indexes of the NAS is also important. In this case, the correlation determines the degree of correlation between the interrelated factors and the resulting factor, whereas regression factors determine how factors are functionally related to the outcome factor and how these factors are effective. In the research, as a resulting factor GDP and influencing factors, expenditure for final consumption-X_1, investment for the main capital-X_2, changes in reserves-X_3 and export-import balances of goods and services-X_4 were selected. The correlation between the resulting factor and the influencing factors was examined. The result is reflected in Table 1 below. Table 3.2 shows that there is a strong connection between the gross domestic product (GDP) and all factors other than the export-import balance of goods and services (0,316028), including final consumption (0,998817), investment in the main capital (0,99954) and changes in reserves (0,86997). It should be noted that multicollinearity did not exist when |r_(〖x_(1) x〗_2) |<0,8 terms of selected factors were performed. This shows that it is possible to ensure the participation of all factors except the exportimport balance of services in the construction of the regression equation, which represents the observed process.
To construct a regression equation, it is appropriate to use the program Eviews, which is currently the most convenient. Accordingly, the following regression equation was determined using the statistical indexes of the dynamics of change of the indexes in 1997-2019 selected from the above: = 664,02 + 0,319 * 1 + 2,86 * 2 + 0,722 3 (1) where: -the amount of gross domestic production; 1expenditures of final consumption; 2investment in the main capital; 3changes in reserves; In this case, of course, it is necessary to check the reliability and adequacy of the identified regression equations in terms of criteria (Table 2). According to the criteria «Akaike», «Schwarz» and «Hannan-Quinn» identified in the evaluation of model (1), the model can be said to be reliable, but it is advisable to eliminate the misunderstanding of the criteria of tstatics given in Table 2. The significance of the parameter of the equation determined by the fact that the value of the degree of significance 05 , 0 =  and the degree of freedom = 21 in the table on the t-criterion of the student presentation is equal to = 2,0796, 1 = 1,042, 1 < , should be checked against the criteria for determining the quality of the forecast model. Since the change in GDP determines TIC = 0,0072 and the quality of the forecast model is MAPE < 10% and 0 ≤ TIC ≤ 1, MAPE = 6,468 < 10%, as well as in the process under consideration, the forecast quality is very high, and the equation (1) -regression is reliable and adequate.
According to the defined model values, the change in cost and reserves per unit of final consumption leads to a decrease in GDP by 0,32 units and an increase of 0,72 units. As a result of this process, it was found that the factor that has the potential to increase GDP relative to all factors is investment in capital, and if this factor is increased by one unit, GDP can be increased by an additional 2.9 units.
According to the analysis, among the indexes of the selected factor, there is multicollinearity in all except the export-import balance of goods and services. The correlation between final consumption and investment in fixed assets is 0,998, and the correlation between investment in fixed assets and changes in reserves is 0,865. The development of industries is expressed by the following exponential equations: where: Y indus. Additionally, the share of industries in GDP was approximated by the following logistics function [13]: Yst-initial cost (t = t0), a-constant amount, was determined according to the given initial term; b -diffusion coefficient was determined using "technology addition" in traditional industries.
In the research, the Eviews program was used to construct a regression equation. Accordingly, the following regression equations were constructed using the statistical indexes of the dynamics of change of the indexes selected above in 1991-2019. The reliability and adequacy of the identified regression equations were tested on the basis of the criteria (Table 3). If we observe the change in GDP in relation to households− 1 , government administration bodies-2 , noncommerce organizations that serve households -3 , gross reserves-4 , we will certainly have a change on some sides. This, of course, reflects the impact of selected factors on GDP [14].  From the data in Table 3, it can be seen that the factors influencing GDP were selected correctly. According to it: 1 = 712.107 + 2,772 * 1 − 7,211 * 2 − 9,325 * 3 + 3,003 * 4 (4) The normalized regression equation emerged. where: 1households; 2 -government administration bodies; 3noncommerce (social) organizations that serve households; 4 -gross reserves.
Surely, it is necessary to check the reliability and adequacy of the identified (2) model based on the criteria, and this is done through the EVIEWS 9 program. The results are shown in Table 5 below. The results of Table 5 show that the parameter 1 was found to be insignificant according to the 1 < . term. However, it is advisable to check whether this parameter is significant or insignificant with another criterion indicator MAPE < 10% and 0 ≤ TIC ≤ 1, which defines the quality of the forecast model and the criteria.
According to the results of testing the (2) model with MAPE = 0,53335 < 10% and 0 ≤ 0.0126 ≤ 1 criteria, which define predictive quality, all the parameters selected for the identified model are important, and the (2) model can be called reliable and adequate.
The model explains that if gross savings are increased by one percent, GDP will increase by 3.0 percent. If expenditures on households, government administration bodies, and noncommerce (social) organizations serve to households increase by one percent, the amount of GDP will decrease by 2,8 percent, 7,2 percent, 9,3 percent.
Multifactor forecast of gross domestic product in model (1) The multifactor forecast of GDP by the regression equations stated above is given in Tables 6-7. Table 4. GDP of the Republic of Uzbekistan and its forecast of sectoral structure for 2019-2025 (billion sums).
Year Industry (together with the field of construction) Agriculture, forestry and fishing industry Services Real export GDP Based on the analytical capabilities of the NAS, it is important to ensure that the national accounting system should consider the methodological principles of accounting for all stages of macroeconomic indexes (beginning with obtaining primary data and ending with the construction of balance tables). It is also necessary to formulate macroeconomic equations in the concept of NAS, the relationship between monetary sectors, the balance of payments and production with the state budget, the distribution of income, redistribution and balance of use.

Conclusions
In this article, based on the study of theoretical and methodological aspects of the statistical analysis of macroeconomic indexes and the international comparison and forecasting of the main indexes of the national accounting system of the Republic of Uzbekistan, the following conclusions have been made, problems have been defined, and scientific recommendations and suggestions have been made. 1.
The research examines the role of national statistics in the economy, the application of the law on statistics in the field, the reforms carried out in recent years to improve the system, the implementation of state programs on the transition to international statistics, and the assessment of macroeconomic indexes in the national system.

2.
The construction of national accounts primarily depends on the quality of the data used to compose them. Often, errors in national accounts are the result of incomplete data or little data. Thus, to ensure the completeness of statistical information, the basic tasks of statistical services should be direct to create a system of a new type of index using new tips and concepts to adapt the data collection and processing tips in accordance with international standards.

3.
Gross domestic product is one of the main indexes of NAS, which represents the final result of production activities of resident economic units and is measured by the value of goods and services produced by these units for final use. GDP is widely used in international practice, and it serves as an indicator of the final results of economic production activities in international and national practice. The main content of this indicator is directly determined by the cost of all final goods and services produced and used (consumed) in the