Econometric studies in forecasting increased yield productivity of agricultural crops

. The article, based on econometric research, examines the relevance of econometric methods of planning and forecasting, as well as methods for compiling econometric models for quantitative analysis and forecasting of agricultural crop yields. In addition, using methods of analysis and forecasting of econometric modeling based on experimental data, the work examined the trend in the dynamics of agricultural crop yields and made a forecast for the future using compiled linear trend equations, as well as provided feedback and suggestions on the results of the study.


Introduction
The five priority directions of the development strategy of the Republic of Uzbekistan for 2017-2021 show the need for modernization and intensive development of agriculture, i.e. deepening structural transformations and the gradual development of agricultural production, further strengthening the country's food security, expanding the production of environmentally friendly products, significantly increasing the export potential of the agricultural sector, reducing cotton and grain crops, planting potatoes, vegetables, fodder and lubricant crops, clearing land, as well as further optimization of cultivated areas through intensive gardens and placement of vineyards, creation of favorable conditions for the encouragement and development of diversified farms engaged in processing, procurement, storage, sales, construction and provision of services, further increasing the competitiveness of our economy in the modern market, modernizing and actively diversifying it industries and networks.
As the President emphasized, we have every opportunity to double the gross domestic product by 2030, and the priority task for further agricultural reform is, first of all, the rational use of land and water resources.
In this regard, special attention will be paid to the widespread introduction of modern intensive agricultural technologies and improvement of the infrastructure for deep processing and storage of products [1][2][3][4][5][6].In our country, we are working to reduce the area sown with cotton in areas with low yields, expand vegetable, legume and food crops using international best practices, and create intensive orchards and vineyards based on high production technologies.

Materials and methods
Research methods: analytical reviews, comparative analysis, experimental testing.
Level of implementation of research results: On the theme of the study, the accumulated experience and publications in the field of information management in the farm management system both in our country and abroad were studied and analyzed.
The first electronic information system "Farms of the Republic of Uzbekistan" was created.
Scope of scientific research: agricultural sector of the Republic of Uzbekistan.

Results
Currently, the main focus will be on expanding export-oriented production of fruits and vegetables and food products, increasing the production of livestock products, as well as organizing small production points and providing services.One of our main tasks is to improve the existing agricultural financing system.It is known that without these efforts it is impossible to ensure stable productivity, development of the entire agro-industrial complex and, most importantly, increase the material interests of farmers and improve the standard of living in rural areas.The implementation of these activities is one of our priorities.Scientific research in this area requires not only the ability to analyze complex socioeconomic processes for a developing country's economy, but also, on the basis of modern econometric, economic and mathematical methods and models with the help of computer technology, determines the increase in demand for these processes in the field of economics and mathematical modeling, which will make multidimensional decisions [7][8][9].This allows you to study trends in the economy, the state of the objects under study and predict their development, and also allows to effectively use limited resources in the national economy to make science-based decisions.Econometric and economic-mathematical methods determine the directions for a comprehensive analysis of all sectors of agriculture, the creation of multi-model models of sustainable economic growth, modernization of agriculture and optimal use of limited resources [7][8][9].
Econometric methods express the problem posed not only analytically, but also help to model the business processes of sectors of the national economy that can determine the influence of endogenous and exogenous factors, creates models that can manage and predict these processes in terms of quality and quantity, and makes it possible to provide advisory proposals or management decisions based on research for the effective management of the object under study.It is well known that the development of computer systems and special applications and the improvement of analytical methods have turned econometrics into a powerful tool for economic research.In any practical feasibility study, econometric methods can be used as the main part of the scientific tool.Therefore, the use of econometric methods is of great importance in the analysis and forecasting of various economic and technical issues.
Forecasting the production of agricultural products in the agro-industrial complex may be a special study that determines the results of the development of directions, objects and phenomena, as well as the prospects for the development of the object.The forecasting process begins with an analysis of the object.This analysis includes the study of factors for object selection that affect the object, its structure and methods of managing it.To solve problems and tasks to increase agricultural productivity, an econometric forecasting model is compiled that links results and factors and is solved with the help of econometric methods using information technology, and optimal solutions are analyzed and predicted for the future.Moreover, the productivity of crops will depend on many factors such as quality of crops, annual rainfall, fertilizers and its types, fertilizer rates, agronomic practices, harvest time, etc. Therefore, there is a relationship between the amount of fertilizers for sown area and yield indicators.Therefore, it is necessary to conduct numerous experiments and observations to determine the indicators that determine the impact on fertility in specific fields.
Creating or modeling a final factor model for analyzing indicators of economic efficiency of an enterprise based on a qualitative analysis of the essence of the economic phenomenon expressed in this indicator is formal and intuitive.Modeling of the indicator model is based on the following economic criteria when selecting factors that are elements of the factorized system: its relevance, its own originality, accessibility and accountability.
In this paper, with the help of analysis methods and forecasting of econometric modeling using the example of three agricultural crops, the trend in the dynamics of agricultural crop yields was studied based on experimental data.The research carries out data processing and econometric analysis of agricultural crop yields (cotton, wheat, potatoes) in the Republic of Uzbekistan for the observation period (2003-2018), as a time series (Table 1).The annual repetition of crop cultivation gives reason to consider this process as a discrete time series , where t-years, -crop yield which consists of trend, fluctuation around trend, seasonality effect and random components.
Based on the statistical data given in Table 1, we process the observed data and analyze crop yields as a discrete time series [10][11][12].The geometric image of the observed data on the yield of agricultural crops (cotton, wheat, potatoes) gives reason to assume the hypothesis that the trend part of the process under study has a linear relationship (Figure 1), y i (t)=a 1 t i +a 0 , where the parameters a 0 , a 1 are determined by the least squares method (LSM), i.e. based on observed experimental data, solving a system of normal equations.Solving the system of normal equations with least squares method, we find the values of the parameters of the linear regression equation and then we obtain linear trends in cotton productivity -, wheat -and potatoes -.  2).In addition, a test of normality showed that the average yield of agricultural crops ӯi (t) has a normal distribution, with the following parameters: cotton -N (24.64; 1.76), wheat -N (42.26; 3.89), potatoes -N(140.018;48.159).Now, based on the observed and obtained forecast results, as well as regression equations for agricultural productivity crops, graphs of linear trend equations are depicted (Figure 2).

Discussion
Analysis of statistical data shows that currently, thanks to the attention and practical assistance provided by the state, the share of agricultural production is growing very quickly.Moreover, the effective development of production processes and the yield of agricultural crops depends on many factors, such as the quality of seeds, annual precipitation, fertilizers and its types, fertilizer rates, agrotechnical measures, harvest time and other factors.Certain amounts of mineral fertilizers that are cultivated in agricultural fields have different effects on the growth of agricultural crop yields in different areas.
Consequently, there is a relationship between the amount of fertilizer for the sown area and yield indicators.This makes it possible to conduct numerous experiments and observations to determine the indicators that determine the effect on the fertility of certain types of fertilizers in specific fields.
Many economic processes and phenomena in agriculture have a correlation, usually in the form of production functions.In this case, the function under consideration is a mathematical model of multifactor economic processes.However, this model represents aspects of the interdependence of events, and it determines what indicators can be obtained as a result of factors influencing production processes.The influence of a large number of factors on production results is studied on the basis of correlation analysis.Correlation analysis is a set of styles of mathematical statistics that illustrates the relationship between the number of connections between the phenomena under.

Conclusion
Thus, the development of production of agricultural crops in the agro-industrial complex, analysis and forecasting of yields is carried out on the basis of correlation analysis and production functions of mathematical statistics.Forecasting based on econometric research, being the basis for scientifically based planning decisions, increasing the scientific level of planning and as one of the foundations of scientific knowledge base methodologies, should serve as a tool to support the concept of medium-and long-term planning, systematic analysis and optimal decision-making.
The implementation of scientific forecasts in practice determines the most effective ways of targeted development of enterprises, and also determines negative trends in economic growth and the most optimal way to use resources, is used to scientifically substantiate the quality of land resources, will help increase crop yields and improve product quality.

Fig. 1 .
Fig. 1.Histogram of agricultural crop yields distribution.Substituting the value t=3.7 into all equations of linear trends, we find the expected forecast of agricultural crop yields on average for the republic in 2021: = 26.38 c/ha, = 48.71c/ha, =180.3 c/ha, and in 2025: 28.7 c/ha, 54.55 c/ha, 208.26 c/ha.Now, to check the presence of autocorrelation in the series of dynamics of agricultural crop yields, using the formula of the Durbin-Watson criterion, we will ascertain that the yield of agricultural crops has an autocorrelation dependence, that is, the yield from agricultural crops in the current year depends on the harvest of the previous year.Based on sample data, we calculate the numerical characteristics of the regression equations

Table 2 .
Numerical characteristics of regression equations.