data150-bree

PVAR

Taking Southwest China as a whole as the research object, this paper selects the panel data of Southwest China from 2000 to 2016 to establish the panel vector autoregressive model (PVAR) model, combined with impulse response and variance decomposition to explore the dynamic relationship among economic development, tourism industry investment and industrial structure optimization, in order to get an objective judgment and provide reference for the formulation of public policy in Southwest China.

As the three core departments of the tourism industry, tourist hotels, scenic spots and travel agencies are the core carriers of regional tourism reception and the key areas of tourism investment, which gather the capital, talent, technology, brand and other resources of regional tourism investment.The most direct manifestation of the optimization and upgrading of industrial structure is that the proportion of the primary industry in the GDP has decreased, while the proportion of the secondary and tertiary industries has increased. Considering that the overall industrial structure optimization of Southwest China is at a low level, and the industrial structure of Tibet Autonomous Region, Guizhou Province and Yunnan Province is gradually transforming from the primary industry to the secondary and tertiary industries, the proportion of non-agricultural output value to GDP is adopted to reflect the industrial structure optimization level. Considering the reliability and availability of data, GDP is selected as the index to measure the regional economic development. The relevant data are from the statistical yearbook, statistical bulletin of national economic and social development, government work report and China Tourism Statistical Yearbook of the provinces in Southwest China. For missing and abnormal values, multiple interpolation methods and regression interpolation methods are adopted to improve the data.

Yit-j is the J-order lag term of Yit, that is, the lag term of an endogenous variable is the explanatory variable. I is the province in Southwest China, t is the year, P is the lag order, η Is the individual effect vector, γ T is the time effect vector, β J is the coefficient matrix, μ i. T is the random interference term.

In order to avoid the phenomenon of pseudo regression in PVAR model and ensure the stability of impulse response and variance decomposition, we use stata15.0 software to test the panel data with LLC (homogeneous unit root) test method and iPS (heterogeneous unit root) test method. If the panel data pass the test, it is considered that the panel data is stable; Otherwise, it is considered that the panel data is not stable. In order to eliminate the possible heteroscedasticity phenomenon in the series, logarithmic processing is adopted for tourism industry investment, economic development and industrial structure optimization respectively, and the cointegrationrelationship of the original data series will not change, which is recorded as lnti, lnGDP and lnind respectively. In the horizontal state, the P values of lnti, lnGDP and lnind do not pass the significance test at the 5% level, and the original hypothesis is not rejected, that is, the panel data is not stable. Ln Ti, lnGDP and lnind are tested for stationarity after the first-order difference, which are respectively recorded as lnTIi, lnGDP and lnIND. The test results all pass the significance test under the 5% level, and the sequence is stable. Therefore, the tourism industry investment, economic development and industrial structure optimization sequence are all single integers.

The response level of economic development to the optimization of industrial structure increased rapidly from the current period to the first period, slowed down and reached the highest value in the second period, and then gradually decreased, but still maintained at a high level, indicating that the optimization of industrial structure has a significant and lasting positive role in promoting economic development.

Panel Data Vector Autoregression, after Holta Eakin combined the VAR model with panel data, it was first put forward in 1988. After the development of many scholars, it has become a very mature model. Compared with an ordinary VAR model, the PVAR model requires less time series length.

GMM Method

Lack of financial resources has always been the core issue of economic development and survival of poor areas and groups in China. Poverty alleviation by means of Inclusive Finance is the key to the implementation of targeted poverty alleviation in China, highlighting the usefulness of finance to vulnerable groups. In 2005, the concept of Inclusive Finance was put forward, which means that under the premise of controllable economic cost, it is no longer limited to large customers and provides relevant services for various groups and different social strata with financial service demand. Low income groups and small and micro enterprises in urban and rural areas are the focus of attention. The definition of Inclusive Finance in domestic theoretical and practical circles is mainly divided into two dimensions: one is that the cost of capital should be low, the traditional financial industry only pays attention to the rate of return on capital, takes high returns as the criterion, and prefers large projects and large companies. When providing financial services to the poor groups and areas, the risk aversion is overemphasized, which artificially increases the financing cost and loan interest, resulting in a serious shortage of financial resources available to these groups and areas.

In Econometrics and statistics, generalized method of moment (GMM) is a general method to estimate the parameters of statistical models. Generally, it is suitable for semi parametric models,in which the parameters of interest are finite dimensional, and the complete shape of the data distribution function may be unknown, so the maximum likelihood estimation is not suitable. This method requires that a certain number of torque conditions be specified for the model. These moment conditions are functions of model parameters and data, so their expected values are zero at the true values of parameters. Then, the GMM method minimizes a norm of sample mean under moment condition, so it can be considered as a special case of minimum distance estimation.

Taking 31 provinces, autonomous regions and municipalities in China as samples, this paper examines the impact of Inclusive Finance on poverty and income inequality. This paper argues that the relationship among Inclusive Finance, income distribution and poverty is not one-way, and there may be reverse causality. For example, when poverty alleviation leads to an increase in demand for banking services, or when economic growth leads to a significant decrease in income inequality, this will lead to an increase in economic demand pressure for more inclusive financial policies. This endogeneity will lead to the potential deviation of the estimated coefficients. In order to overcome this potential deviation.

Where, φ、γ、β all coefficients are to be estimated. The subscript i denotes the country, T is the time. Xit is the core explanatory variable, which is the proxy variable of Inclusive Finance, and Yit is the other economic variable, which is included in the model as the control variable. Based on the purpose of this paper to examine the impact of Inclusive Finance on poverty and income inequality, the dependent variable zit on the left of the above equation represents the poverty and income inequality of the ith country in the T period. Zit-1 on the right side of the equation is the lag period of the dependent variable, which reflects the dynamic nature of the model. α I is a fixed effect, which controls the unobservable variables at the national level. ε It is the error term of an econometric model, which is independent and identically distributed in the whole sample.

Among the control variables, the lag term of income inequality (lag term) is the most important_ The sign of equality is not consistent, positive and negative, but not significant. The enrollment rate is consistent . A higher enrollment rate in secondary schools will restrain the widening of the income gap. The coefficient of the proportion of import and export in GDP (openness) is negative, but not significant. Inflation rate has significantly increased the degree of income inequality. The possible reason for inflation to widen the gap between the rich and the poor is that the investment decisions of the rich and the poor are different. When inflation occurs, their real wealth will be further enlarged. In the GM model, GDP per capita is significantly positive, which indicates that the increase of general per capita income will reduce the further widening of income inequality.

In this way, in southwest China, inclusive finance can increase the income of specific groups (poverty alleviation objects) through the availability of Finance and reducing the cost of financial products and services, promoting finance to ease credit constraints and play a role in preventing risk shocks. The main mechanisms are as follows: firstly, by increasing the coverage of commercial banks and ATMs, we can improve the availability of financial services for the poor. Secondly, strengthen the policy guidance of commercial banks, and enhance the ability of financial product innovation, greatly alleviate the credit constraints of poor groups and poor areas.