On Farmer’s Economic Income in Hubei Province of China During the COVID-19 Epidemic

This paper discusses the statistical measurement of the impact of COVID-19 major emergencies on farmers’ economic income in Hubei Province. Hubei Province was selected as the object of analysis, and five data of total output value of agriculture, forestry, animal husbandry, fishery and per capita disposable income of farmers in Hubei Province from the first quarter of 2013 to the second quarter of 2020 were collected by using the Internet. Since all the collected data were macroeconomic data, these data were taken the logarithm to meet the economic significance. The per capita disposable income of farmers was taken as the response variable, and the main factors affecting farmers’ income were obtained by factor analysis. Livestock husbandry and fishery industries were the main industries in Hubei Province. Then the score of factor analysis were taken as explained variable to establish regression model composed of influencing factors. This paper use the multiple linear regression, support vector regression to fitting and forecasting data, ARIMA model of time series analysis, introduced at the same time, through the AIC model choice, with the first quarter of 2013 to 2019 in the second quarter fitting training, backward prediction two quarters, and three or four quarter of 2019 compared with the real data, through to the predicted results of the sequence diagram and evaluation index model to compare the mean square error (RMSE). Three models predict per capita disposable income of farmers in the first and second quarter of 2020. It has been found that performance better ARIMA model in the model compare is worse than before, and three kinds of predicted values are higher than the real value of the model, showed the outbreak to the influence of the agricultural economy in hubei province is serious. On this basis, taking into account the characteristics of geomorphic climate in Hubei province, the constructive suggestions are put forward. (Known-Unknowns),


Introduction
Thanks to the hard work of the whole nation, the epidemic prevention situation in China has been constantly improved, and the order of production and life has been quickly restored. In order to win the critical period of building a well-off society in all aspects and the decisive battle against poverty, the impact of COVID-19 epidemic on agriculture and rural economy should be scientifically studied and judged, and appropriate responses should be made. It is of great significance to ensure the well-being of a well-off society and the quality of poverty alleviation. The impact of COVID-19 on agricultural economy mainly includes the impact on the supply of agricultural products, the impact on livestock, poultry and aquaculture, the impact on the supply of agricultural materials and spring tillage preparation, and the impact on the international trade of agricultural products, etc [1,2].
Research on the relationship between agricultural industrial structure and economic growth from the perspective of farmers' income level: Chen Kai [3] used the grey correlation method to analyze the relationship between rural household operating income and the output value of planting, forestry, animal husbandry and fishery in the Yangtze River Delta region. Yang Zhong-Na et al. [4] analyzed the impact of agricultural structure and its changes on agricultural growth in southern Xinjiang from 2000 to 2011 and pointed out that agricultural industrial structure has restricted agricultural economic growth, indicating that the relationship between agricultural industrial structure and agricultural economic growth is spatially regional [5,6]. Therefore, it is very important to analyze the impact of the epidemic on agricultural economy, which can guide the direction of agricultural structural reform [7]. Figure 1: Partial data diagram.

Index Selection
In this paper, the main income sources of farmers are selected as the analysis indicators, including agriculture, forestry, animal husbandry and fishery. Some data are shown in Figure 1. First, taking logarithm satisfies the economic significance, and then, through factor analysis, the internal relations between various variables are explored to mine the internal structure of the observed data and provide convenience for subsequent professional analysis.

Establishment and Solution of the Model
The Flow Chart 1 of the overall thinking is as follows: Feasibility of factor analysis of data: KMO test and Barrett ball test were performed on the data when the KMO value was greater than 0.5. When P-value of Bartlett test is less than 0.05, it is considered that factor analysis can be carried out.
iii. Carry out EFA analysis to determine the number of common factors, solve the characteristic equation 0 I R λ − = and solve the eigenvalues, and arrange them in order from large to small to draw parallel gravel diagram. The current n eigenvalues are above the corner, and they are all greater than the mean value of the eigenvalues of the simulation data matrix for 100 times, and the number of common factors can be determined as n.
iv. Factor rotation: The oblique rotation is adopted to extract the factor, which is beneficial to better explain the practical significance of the factor.  Factor Analysis: Spearman correlation coefficient analysis of pre-data calculation in Hubei province was conducted in this paper, and the correlation coefficient matrix was visualized. The results are shown in Figure 2. As can be seen from Figure 2, there is a strong correlation between all variables, with a correlation coefficient above 0.8 showing a strong correlation. Therefore, factor analysis is considered in this paper to eliminate the correlation between variables. KMO test and Bartlett spherical test were performed on the data, and the test results were shown in Table 1. As can be seen from Table 1, the KMO value of the data test is greater than 0.5, and the P value of bartlett's spherical test is less than 0.05. Therefore, the test results can be considered as applicable to factor analysis. The R programming language was used to draw a parallel gravel diagram ( Figure 3) to determine the number of common factors.
In Figure 3, we drew the gravel diagram for principal component analysis for comparison. If PCA principal component analysis was used in the diagram, the number of principal components to be selected from the PC curve was 1. According to the FA curve, the first two eigenvalues are above the corner and are greater than the mean value of the eigenvalues of the simulated data matrix for 100 times. Therefore, according to The Kaiser-Harris principle and the parallel analysis criteria, it can be considered that two common factors can be selected for factor analysis. We determined that it was appropriate to select two common factors, and then carried out factor rotation on the data of Hubei Province, and drew oblique rotation factor result chart to determine the influencing factors represented by each factor. It can be concluded from the Figure 4 that factor 1 represents the driving factor of animal husbandry and factor 2 represents the driving factor of fishery.

Factor Score
The results of the factor score table are shown in Table 2. The corresponding expressions of the two factors can be obtained by converting the data in Table 2 into the factor score function: According to the information in Table 2, the common factor 1 represents the animal husbandry driving factor; Common factor 2 represents the fishery driving factor. According to the above analysis, it can be known that for Hubei Province, the main influencing factors of agricultural structure adjustment on farmers' income include driving the overall economic development led by animal husbandry and vigorously developing animal husbandry, among which the first influencing factor is the most important. The closer R square is to 1, the greater the degree of interpretation of independent variables to dependent variables will be. It is known from the table that the common factors of the multiple regression model have a good interpretation of farmers' income (Table 3). the value of the common factor of the multiple regression model is close to 2, so it can be considered that the common factor has good independence ( Table 4).  there is no multicollinearity between the common factors selected in Section 3.1 (Table 6).

Model prediction
In order to study the impact of the epidemic on agricultural  (Table 7). The time sequence diagram of model comparison and prediction is shown in Figure 5. It can be seen from Figure 5 that the

predicted value in the third and fourth quarters of 2019 in Hubei
Province is lower than the real value, indicating that the farmers' income is rising steadily, while the predicted value in the first and second quarters of 2020 is higher than the real value, indicating that the rural economy in these two quarters is seriously affected by the epidemic.
The objective function of SVR model is: The following Lagrangian function is obtained: re Lagrange coefficients.
Finally, using SMO algorithm to find the corresponding i α ∨ and i α ∧ then we can get the coefficient of our regression model ω , b

Analysis and Interpretation of Model Results
Time sequence diagram of model comparison and prediction is shown in Figure 8. It can be seen from the results that the predicted values in Hubei Province are all higher than the real values, indicating that the income of farmers in Hubei Province is seriously affected by the epidemic.  In addition, we made ACF and PACF graphs of residuals to judge whether the residuals are white noise. If ACF and PACF of each order are less than the critical value of the test, the residual is considered to be white noise, that is, the model identifies the data well.
Otherwise, it is considered that the model is not fully identified. The result is shown in Figure 9. By analyzing the ACF and PACF graphs of the time-series model residuals corresponding to all variables in Hubei Province, it can be found that they are all less than the critical value of the test. Therefore, the residuals can be considered as white noise. That is to say, the model obtained in this paper can identify our data well. So we can accept these temporal models.

Comparative Analysis of Models
The mean square errors of comparison and prediction of all models are shown in Table 8 below. As can be seen from

Conclusions and Recommendations
Based on the modeling and analysis in this paper, we successively put forward suggestions on the impact of agricultural structure adjustment on farmers' income in Hubei Province.

Strengthen Macro-Control
Although we strongly advocate market agriculture and

Develop Aquaculture Vigorously
The low proportion of aquaculture in the primary industry is an important manifestation of the irrational agricultural industrial structure. Hubei province has rich water resources and unique advantages for aquaculture, but the proportion of fishery in the agricultural industry in Hubei province is only higher than forestry.
Therefore, we should pay more attention to aquaculture and vigorously develop aquaculture with local characteristics, such as crayfish in Qianjiang and crab in Xianning, so as to make it the key point to promote farmers' income.

Strengthen Scientific Training
Local governments and agricultural departments should stick to the central idea of "developing agriculture through science and technology", persist in improving farmers' quality through "science and technology training" as a media platform, and help farmers solve various technical problems they may encounter in production, so as to improve farmers' agricultural planting techniques. Through the combination of "centralized teaching, on-site teaching and individual guidance", the innovative mode of agricultural cultivation is promoted, and the prevention and control analysis of possible insect disease disasters is carried out, so that farmers can quickly master skills. The government's science and technology training platform plays an important role in strengthening farmers and radiating the surrounding economic development.

Agricultural Economy
Combined with China's special rural conditions, coupled with the lack of implementation of relevant national benefits to farmers, their own land, lack of funds, single information channels, resulting in farmers in the process of agricultural industrial structure adjustment in the face of the market alone in the case of higher costs, limit the increase of income level. In the face of this situation, it is very important to gradually develop a new form of cooperative organization to help the separated and independent farmers establish connections with the large market.