Original Paper Research on the Impact of GNI on Express Delivery Volume in China

This paper selects the relevant data of 30 years from 1989 to 2018 from the National Bureau of Statistics. This paper selects the gross national income as the focus variable, and the number of business outlets, cargo transportation volume, investment in fixed assets and the total population of the country as the control variables to make an empirical analysis on the influencing factors of express delivery volume in China. EVIEWS software is used to estimate, test and correct the parameters of the model. The economic significance of the final results is analyzed, and then the research conclusion is drawn and the existing deficiencies are summarized.


111
Published by SCHOLINK INC. conducive to clarifying the current social development trend and the changes in people's lifestyle and consumption concepts.

Research Contents and Methods
This paper consists of five chapters. The research framework of this paper is as follows: The first chapter is about the influencing factors of the number of express delivery in China. Firstly, it explains the factors that affect the number of express delivery in China and the focus variable that affects the number of express delivery in China-GNI, including the external factors such as domestic economic development and the development of public investment, etc. On this basis, it selects the control variables.
The second chapter is of data description. It mainly explains the data sources and explains the relevant variables, and describes the characteristics of data change trends, laying a foundation for the next empirical analysis.
The third chapter is an empirical analysis of the influencing factors of China's express delivery quantity. The regression analysis model in econometrics is used for analysis. Based on time series data samples, a time series model is constructed. For the processing of the model, firstly, the unit root test, namely ADF test, is carried out. Then the cointegration test of the explanatory variables and the explained variables in the model is carried out, It is found that there is a cointegration relationship at the significance level of 0.05. Then an error correction model is established to further explore the significance and influence degree of explanatory variables on the explained variables. Finally, Granger causality test is carried out to test whether there is a one-to-one corresponding prediction causality between the explanatory variables and the explained variables.
In the fourth chapter, the adjusted model is tested for heteroscedasticity and autocorrelation, and the cointegration model is tried to be tested to make the model more in line with reality.
The fifth chapter is the research conclusion and deficiency. Firstly, the conclusions drawn from the empirical analysis of the multiple regression model in this paper are summarized, and reasonable suggestions are put forward to promote the stable development of China's logistics express industry. Combined with the model and the actual analysis one by one, the deficiencies in the research process are put forward.

Model Selection
Since the hypothesis testing of nonlinear models involves very complicated mathematical calculations, this paper considers making a linear model, so there are many testing methods and the analysis of the accuracy of the model is more reliable.

Variable Selection
There are many factors that affect the volume of express delivery, including the number of business outlets nationwide, cargo volume, investment in fixed assets, the total population of the whole country, www.scholink.org/ojs/index.php/jepf Journal of Economics and Public Finance Vol. 6, No. 3, 2020 112 Published by SCHOLINK INC.
the gross national income, etc. But considering comprehensively, this paper selects the gross national income as the focus variable, and the number of business outlets, cargo transportation volume, investment in fixed assets, and the total population of the whole country as the control variables to conduct the research. In order to find the data conveniently, this paper selects the relevant data from 1989 to 2018 from the National Bureau of Statistics.

Number of Online Stores
The amount of express delivery is closely related to people's online shopping level. The more business outlets, the more able they are to deliver express delivery, which naturally attracts consumers.
Therefore, the number of business outlets is related to the amount of express delivery. This paper predicts that this factor is positively related to the amount of express delivery.

Cargo Transportation Volume
This paper predicts that the amount of express delivery is related to the amount of goods transported.
The higher the amount of goods transported, the higher the efficiency of express delivery, which also indicates that the country or region has a higher level of economic development and people's consumption concept can keep up with the trend of the times. Therefore, the explanatory variable cargo transportation volume is introduced, and a priori it is expected that it is positively correlated with express delivery volume.

Investment in Fixed Assets
Investment in fixed assets is a comprehensive index that reflects the scale, speed, proportion and use direction of investment in fixed assets. Fixed asset investment has a significant impact on the distribution and development of industries in a region. It can not only stimulate regional economic growth, but also improve public infrastructure and service level for the public. For example, for the part of investment in fixed assets that is used for infrastructure investment, transportation investment can expand the transportation network, greatly shorten the transportation time of express delivery on the road, and improve the efficiency of express delivery. Investment in the postal industry can add outlets of rural postal and express delivery enterprises to improve the level of express delivery services.
Investment in electric wires can speed up the construction of the Internet, expand the coverage of the Internet, provide a broader platform for online shopping, and further promote the increase in the demand for express delivery. This paper predicts that the volume of express delivery is related to GDP, so the explanatory variable GDP is introduced and a priori expects that it is positively related to the volume of express delivery.

Total Population
The number of express delivery refers to the number of goods purchased online by the national population. The larger the population, the more potential consumer groups, the more likely the online shopping is. This paper predicts that express delivery is related to the total population of the country, so the explanatory variable of the total population of the country is introduced, and a priori predicts that it is positively related to express delivery volume.

GNI
This article predicts that the volume of express delivery is related to the gross national income, The higher the gross national income, the higher the level of economic development and people's living standard is. A high level of economic development means a fast-paced life. A high quality of living also means that people are pursuing higher living goals. Online shopping is a new way of shopping and meets people's living needs. Therefore, the explanatory variable total import and export volume is introduced, and a priori it is expected that it is positively correlated with express delivery volume.

Sources and Processing of Data
In this paper, the relevant data of 30 years from 1989 to 2018 from the National Bureau of Statistics are selected and processed: Y indicates the express delivery volume (10,000 pieces); X1 indicates the number of business outlets (number); X2 represents the freight volume (10,000 tons); X3 indicates the investment in fixed assets (100 million yuan), X4 indicates the total population of the country (10,000 people); X5 represents gross national income (billion yuan); N is a random perturbation term. The data are shown in Table 1. According to the data provided in Table 1, the statistical software Eviews8 is used to make a scatter plot for the above-mentioned set model, as shown in the following figure.

Figure 1. Relationship between National Express Delivery Volume and Various Variables
In order to obtain a more accurate multiple regression model and make a more accurate prediction, we first use Eviews software to estimate the linear relationship between Y and Xi. As can be seen from     are selected as explanatory variables. Based on the above analysis, we set up the model as follows:

OLS analysis between Y and Explanatory Variables
Y=β1+β2* X1+β3* X2+β4* X3+β5* X4+β6* X5+u Since all time series data are subject to unit root test, we will carry out unit root test on the preliminarily determined model in the next step. If there is a unit root, we will correct the model. When the unit root test is carried out on X1, p = 0. 9999 is very close to 1, which indicates that there is a unit root. When the second-order difference correction is carried out on X1, given the significance level = 0.05, P = 0. 0022 < 0.05, and t = -4. 9728, H0 is not acceptable, and the sequence has no unit root and is gentle.

Unit Root Inspection and Correction of X1
The Unit Root Test and Correction of Y  Similarly, ADF test is carried out on other control variables to obtain the following table.

Cointegration Test
After ADF test of variables, further co-integration test is carried out on the model to test whether non-stationary explained variables and explanatory variables have co-integration concern. If so, an error correction model is established, the regression equation is estimated by OLS method, and the residual sequence is obtained. ADF test is carried out on the residual sequence. The results are as follows.

Figure 12. ADF Test Results for Residual Sequence
Because of N=6, according to the critical value calculation formula: Therefore, the regression model has a cointegration relationship, indicating that there is a long-term equilibrium concern between the explanatory variable and the explained variable, but in the short term, there may be imbalance. In order to enhance the accuracy of the model, an error correction model is established to link the short-term factor changes with the long-term changes in the number of express delivery in China.
The structure of the modified model is as follows:  After debugging, the above-mentioned lag term (-1) is the best form of short-term relational regression model. The results show that, Under the condition that other variables remain unchanged, the number of business outlets has a significant impact on the number of national express delivery, the volume of goods transported and the investment in fixed assets have a negative impact on the number of national express delivery, and the total population of the country has no significant impact on the number of national express delivery. Under the condition that the confidence level is 100%, it is believed that the gross national income has a significant impact on the number of national express delivery.
The above analysis has proved that there are four factors that affect the number of express delivery in the country: the number of business outlets, the volume of cargo transportation, the investment in fixed assets and the gross national income all have a long-term equilibrium relationship. On this basis, Granger causality is used to further analyze whether there is causality between each explanatory variable and non-explanatory variable and the direction of influence.

Granger Causality Test
Granger's causality test thought is: if the change of X causes the change of Y, the change of X should occur before the change of Y. Granger causality test can only test the causality between two variables.
For the two-variable causality test, the causality between αand λ is judged by checking whether the sum parameters in the following two formulas are all zero.
According to whether the sum parameters are all zero, there are four possibilities for the test results, and two of them are mainly considered in this paper, namely: (1) X has a single influence on Y, which shows that at least one of the parameters before each lag term of formula X is not zero, while all the parameters before each lag term of formula X are zero.
As can be seen from that inspection result, There is a bidirectional Granger causality between the express delivery volume and the freight volume when the lag period is 2, There is a single Granger causality between the freight volume and the investment in fixed assets to the express delivery volume when the lag is 4periods. When there is a lag of 5 periods, all variables have no Granger causality with the explained variables.

Figure 17. Correlation LM Test
The value of the test statistic nR2 is 10.1312, and the 2 distribution table can be found to be 0.052 (6) = 12.59 > 10.1312. Therefore, the original hypothesis cannot be rejected, and it is believed that the model does not have first-order sequence correlation and does not need correction. (1) From the error correction results, it can be seen that the volume of online stores, cargo transportation, total fixed assets, the total population of the country and the gross national income have a significant impact on the volume of express delivery in China. Under the condition that the confidence level is 100%, it is believed that the gross national income has a significant impact on the number of national express delivery, which shows that for every unit of national income increase, the number of express delivery increases by 17.05625 units. The number of business outlets has a www.scholink.org/ojs/index.php/jepf Journal of Economics and Public Finance Vol. 6, No. 3, 2020 123

Research Conclusions
Published by SCHOLINK INC.
significant impact on the number of national express delivery under the confidence level of 74.57%, which shows that for every unit added to the number of business outlets, the number of express delivery increases by 73,883.12 units. Goods transportation volume and fixed asset investment have a negative impact on the national express delivery volume, i.e., For each additional unit of goods transportation volume and fixed asset investment, the express delivery volume will decrease by 1.28 units and 7.24 units respectively. The total population of the whole country has no significant influence on the number of express delivery in the whole country.
(2) From 1989 to 2018, under the condition that other explanatory variables remain unchanged, with China's economic growth, the growth rate of express delivery volume caused by the increase in the number of business outlets is larger than the growth rate of express delivery volume caused by the increase in national income. Therefore, the number of business outlets is the main factor affecting the amount of express delivery volume. That is, the number of business outlets has promoted the infrastructure construction in all regions of the country, increased people's attention and participation in online shopping, and the volume of express delivery has only increased.
( (4) Although this model does not take into account gender, age, the development of e-commerce industry and the educational level of residents, it does not mean these factors have no influence.
Women are the favourites of online shopping. The development of e-commerce has a great impact on the younger generation. The development from computers to mobile terminals such as mobile phones and iPad has made it more convenient for people to buy online, thus affecting the number of express delivery.

Policy Recommendations
(1) Increase the number of business outlets and strengthen the supporting construction of highway transportation infrastructure.
Generally speaking, when people shop online, they will first consider the number of business outlets and the construction of transportation infrastructure in the place where they are located. transportation is in cities and towns, the more willing the residents will have in online shopping, because it not only saves people precious time, but also can experience the excitement and pleasure of shopping without leaving their homes.
(2) Accelerate the transformation and upgrading of the domestic electronic service industry and improve the informatization level of the tertiary industry.
Since the reform and opening up and the development of e-commerce industry, the degree of social openness in our country has improved significantly, but there are still some areas that do not know or even are completely unfamiliar with the Internet and e-service industry. Therefore, popularizing Internet knowledge, improving the structure of electronic service industry and improving the informatization level of the tertiary industry are major issues of the times. In accelerating industrial transformation and upgrading, the upgrading of electronic service industry is as important as the upgrading of energy industry. Entering a new era, people pay more attention to the quality of service and pursue an exquisite life. Popularization of online shopping information and specialization of service are conducive to people enjoying better service experience.
(3) Further strengthen the construction of network security and social security Network security is an important factor that affects people's online shopping experience and thus the number of express deliveries. Therefore, the network supervision department needs to step up efforts to crack down on network fraud ,strengthen the construction of network security, ensure the safety of people's funds, and improve the trust of the network. In addition, social security cannot be ignored, especially public health security. People will only increase their consumption and purchase on the basis of ensuring their life and health. Therefore, relevant departments could establish relevant systems to eliminate potential safety hazards and achieve early reporting, early implementation and early management.

Inadequate Research
In this paper, multiple linear regression is used to predict the demand for express delivery. The model has strong practicability and high accuracy, but the error in the prediction itself cannot be avoided. We attach much importance to the impact of GNI on the number of express deliveries. According to the analysis of economic significance, these five factors are positively correlated with the number of express delivery. However, due to the error of model setting, the influence of missing variables and the choice of correction methods, the symbols of regression coefficients are opposite to expectations. Therefore, the model set in this paper is only for reference and cannot be used as a direct investigation of the relationship between economic variables.