Seeking help with econometrics paper

Econometrics

Final Lab Report

Title of Experiment: Analysis of Per Capita Consumption Expenditure of Urban Residents in Large and Medium-sized Cities and Its Influencing Factors

Surname:

School No.:

Class:

Supervising Instructor:

Time:

Per capita consumption expenditure of urban residents in 23 cities.

Analysis of its influencing factors

I. Background of economic theory

In recent years, China's economy has maintained a rapid development momentum, investment, export, and consumption form the "three horse carriages" driving economic development, which has been recognized by all walks of life***. Through the establishment of an econometric model, the use of econometric analysis of the factors affecting the per capita consumption expenditures of urban residents, to find out the key factors affecting the policy makers to provide a certain reference, and ultimately promote the consumption demand of this "wagon" can become the cornerstone of leading China's economy to a healthy, rapid and sustained development.

The theory of per capita consumption expenditure and its influencing factors

We mainly analyze the influencing factors of China's residents' consumption expenditure from the following aspects:

1) Residents' expectation of future expenditure rises, which affects the growth of residents' immediate consumption

Residents' passive savings directly lead to a huge diversion of purchasing power, thus weakening immediate demand for consumer goods, which seriously affects residents' immediate consumption

The residents' passive savings directly lead to a huge diversion of purchasing power, thus weakening immediate demand for consumer goods, which seriously affects residents' immediate consumption. Since the late 1990s, a series of reform measures in China's medical care, pension, unemployment insurance, education and other reform measures have been introduced, the original system is broken, while the new system has not yet been established, so the current medical care, pension, unemployment insurance, education system on the residents of individual spending pressure is greater, and basically are hard. Pressure is greater, and basically hard spending, spending uncertainty is also great, leading to the current rise in residents' expectations of future spending.

②, the structural contradiction between the supply and demand of goods is still prominent

From the consumption structure, China's consumer goods market has undergone a new fundamental change: residents of the low-level consumption has been nearly saturated, and a higher level of consumption has not been reached. Reform and opening up for more than 20 years, urban and rural residents after a mid-range consumer durables after the popularization stage, the current income of the people's consumption is not enough to form a new, high-grade products as the content of the dominant hotspots of consumption, such as cars, housing, etc. is still far from being able to incorporate into the mainstream of the majority of people's consumption, the residents of the existing purchasing power can not be formed to promote the upgrading of dominant consumer goods drive.

③, the overall price level continues to run at a low level, deflationary pressure is greater, is not conducive to the growth of consumption

After joining the WTO, with the reduction of tariffs and the expansion of the scale of imports, the impact of foreign products on China's market will be further increased, and the international price crunch will have a negative impact on the domestic price changes. The continued decline in prices is not conducive to the growth of consumption of residents. Because from the residents of the consumer psychology, buy up not buy down is the habit of residents shopping psychology. Since residents have expectations of further price declines, they tend to postpone consumption, which is not conducive to the growth of residents' consumption. In addition, statistically analyzed, due to the decline in prices, nominal consumption growth is often lower than the growth of real consumption, which to a certain extent is also not conducive to the increase in the rate of consumption growth.

4, China at this stage did not form a large consumption hotspot, it is difficult to drive the rapid growth of consumption

After the cultivation and development in recent years, China has formed a housing consumption, residential automobile consumption, consumption of communications and electronic products, holiday consumption and tourism consumption and some other consumption highlights, which can promote the stable growth of consumption, but has not been able to form a large consumption hotspot

There are some consumption highlights that can promote the stable growth of consumption, but they have never been able to form big consumption hotspots, and therefore cannot drive the high speed growth of consumption.

Third, relevant data collection

Related data are all from the 2006 China Statistical Yearbook:

Basic information on urban households in 23 large and medium-sized cities

Region Average number of people employed per household (persons) Average number of people burdened by each employed person (persons) Average monthly income per person in real terms (yuan) Disposable income per capita (yuan) Per capita consumption expenditure (yuan) (yuan)

Beijing 1.6 1.8 1865.1 1633.2 1187.9

Tianjin 1.4 2.0 2010.6 1889.8 939.8

Shijiazhuang 1.4 2.0 1061.3 1010.0 722.9

Taiyuan 1.3 2.2 1256.9 1159.9 789.5

Hohhot 1.5 1.9 1354.2 1279.8 772.7

Shenyang 1.3 2.1 1148.5 1048.7 812.1

Dalian 1.6 1.8 1269.8 1133.1 946.5

Changchun 1.8 1.7 1156.1 1016.1 690.2

Harbin 1.4 2.0 992.8 942.5 727.4

Shanghai 1.6 1.9 1884.0 1686.1 1505.3

Nanjing 1.4 2.0 1536.4 1394.0 920.6

Hangzhou 1.5 1.9 1695.0 1464.9 1264.2

Ningbo 1.5 1.8 1759.4 1543.2 1271.4

Hefei 1.6 1.8 1042.5 950.1 686.9

Fuzhou 1.7 1.9 1172.5 1059.4 942.8

Xiamen 1.5 1.9 1631.7 1394.3 998.7

Nanchang 1.4 1.8 1405.0 1321.1 665.4

Jinan 1.7 1.7 1491.3 1356.8 1071.4

Qingdao 1.6 1.8 1495.6 1378.5 1020.7

Zhengzhou 1.4 2.1 1012.2 954.2 750.3

Wuhan 1.5 2.0 1052.5 972.2 853.1

Changsha 1.4 2.1 1256.9 1148.9 986.8

Guangzhou 1.7 1.8 1898.6 1591.1 1215.1

Fourth, the establishment of the model

According to the data, we establish the general model of multiple linear regression equation as:

Where:

- per capita consumption expenditure

- constant term

- -parameters of the regression equation

-average number of employed persons per household

-average number of persons burdened per employed person

- -Average real monthly income per person

-Disposable income per capita

-Following error term

V. Experimental Procedure

(I) Regression model Parameter estimation

Based on the data to establish multiple linear regression equations:

First of all, using Eviews software for OLS estimation of the model, to get the sample regression equation.

Using Eviews the output results are as follows:

Dependent Variable: Y

Method: Least Squares

Date: 12/11/07 Time: 16:08

Sample: 1 23

Included observations: 23

Variable Coefficient Std. Error t-Statistic Prob.

C -1682.180 1311.506 -1.282633 0.2159

X1 564.3490 395.2332 1.427889 0.1704

X2 569.1209 379.7866 1.498528 0.1513

X3 1.552510 0.629371 2.466766 0.0239

X4 -1.180652 0.742107 -1.590947 0.1290

R-squared 0.721234 Mean dependent var 945.2913

Adjusted R-squared 0.659286 S.D. dependent var 224.1711

S.E. of regression 130.8502 Akaike info criterion 12.77564

Sum squared resid 308191.9 Schwarz criterion 13.02249

Log likelihood - 141.9199 F-statistic 11.64259

Durbin-Watson stat 2.047936 Prob(F-statistic) 0.000076

According to the multivariate linear regression about the Eviews outputs one can get the estimation of the parameters as follows: , , , ,

Thus the preliminary regression equation is obtained as:

Se= (1311.506) (395.2332) (379.7866) (0.629371) (0.742107)

T= (-1.282633) (1.427889) (1.498528) (2.466766) (- 1.590947)

F=11.64259 df=18

Model Test: Since the p-value of the test of the explanatory variables , , is greater than 0.05 at the level of , the variables are not significant, which indicates that there may be problems such as multiple **** linearity in the model and thus the model is corrected.

(II) Dealing with Multiple **** Linearity

We use stepwise regression to test and deal with the multiple **** linearity of the model:

X1:

Dependent Variable: Y

Method: Least Squares

Date: 12 /11/07 Time: 16:28

Sample: 1 23

Included observations: 23

Variable Coefficient Std. Error t-Statistic Prob.

C 153.8238 518.6688 0.296574 0.7697

X1 523.0964 341.4840 1.531833 0.1405

R-squared 0.100508 Mean dependent var 945.2913

Adjusted R- squared 0.057675 S.D. dependent var 224.1711

S.E. of regression 217.6105 Akaike info criterion 13.68623

Sum squared resid 994441.2 Schwarz criterion 13.78497

Log likelihood -155.3917 F-statistic 2.346511

Durbin-Watson stat 1.770750 Prob(F-statistic) 0.140491< /p>

X2:

Dependent Variable: Y

Method: Least Squares

Date: 12/11/07 Time: 16:29

Sample: 1 23

Included observations: 23

Variable Coefficient Std. Error t-Statistic Prob.

C 1756.641 667.2658 2.632596 0.0156

X2 -424.1146 347.9597 -1.218861 0.2364

R-squared 0.066070 Mean dependent var 945.2913

Adjusted R-squared 0.021597 S.D. dependent var 224.1711

S.E. of regression 221.7371 Akaike info criterion 13.72380

Sum squared resid 1032515. Schwarz criterion 13.82254

Log likelihood -155.8237 F-statistic 1.485623

Durbin-Watson stat 1.887292 Prob(F-statistic) 0.236412

X3:

Dependent Variable: Y

Method. Least Squares

Date: 12/11/07 Time: 16:29

Sample: 1 23

Included observations: 23

Variable Coefficient Std. Error t- Statistic Prob.

C 182.8827 137.8342 1.326831 0.1988

X3 0.540400 0.095343 5.667960 0.0000

R-squared 0.604712 Mean dependent var 945.2913

Adjusted R-squared 0.585888 S.D. dependent var 224.1711

S.E. of regression 144.2575 Akaike info criterion 12.86402

< p>Sum squared resid 437014.5 Schwarz criterion 12.96276

Log likelihood -145.9362 F-statistic 32.12577

Durbin-Watson stat 2.064743 Prob(F-statistic) 0.000013

X4:

Dependent Variable: Y

Method: Least Squares

Date: 12/11/07 Time: 16:30

Sample : 1 23

Included observations: 23

Variable Coefficient Std. Error t-Statistic Prob.

C 184.7094 161.8178 1.141465 0.2665

X4 0.596476 0.124231 4.801338 0.0001

R-squared 0.523300 Mean dependent var 945.2913

Adjusted R-squared 0.500600 S.D. dependent var 224.1711

S.E. of regression 158.4178 Akaike info criterion 13.05129

Sum squared resid 527020.1 Schwarz criterion 13.15003

Log likelihood -148.0898 F-statistic 23.05284

Durbin-Watson stat 2.037087 Prob(F-statistic) 0.000096

From the resulting data, it can be seen that the adjusted coefficient of determination is the largest, and therefore firstly introduce the The adjusted equation is introduced into the equation, and then the variables , , are introduced into the equation to carry out OLS:

X1, X3

Dependent Variable: Y

Method: Least Squares

Date: 12/11/07 Time: 16:32

Sample: 1 23

Included observations: 23

Variable Coefficient Std. Error t-Statistic Prob.

C -222.8991 345.9081 -0.644388 0.5266

X1 289.8101 227.2070 1.275533 0.2167

X3 0.517213 0.095693 5.404899 0.0000

R-squared 0.634449 Mean dependent var 945.2913

Adjusted R-squared 0.597894 S.D. dependent var 224.1711

S.E. of regression 142.1510 Akaike info criterion 12.87276

Sum squared resid 404138.2 Schwarz criterion 13.02087

Log likelihood -145.0368 F-statistic 17.35596

Durbin-Watson stat 2.032110 Prob(F-statistic) 0.000043

X2, X3

Dependent Variable: Y

Method: Least Squares

Date: 12/11/07 Time: 16:33

Sample: 1 23

Included observations: 23

Variable Coefficient Std. Error t-Statistic Prob.

C 239.5536 531.1435 0.451015 0.6568<

X2 -27.00981 244.0392 -0.110678 0.9130

X3 0.536856 0.102783 5.223221 0.0000

R-squared 0.604954 Mean dependent var 945.2913

Adjusted R-squared 0.565449 S.D. dependent var 224.1711

S.E. of regression 147.7747 Akaike info criterion 12.95036

Sum squared resid 436747.0 Schwarz criterion 13.09847

Log likelihood -145.9292 F-statistic 15.31348

Durbin-Watson stat 2.063247 Prob(F- statistic) 0.000093

X3, X4

Dependent Variable: Y

Method: Least Squares

Date: 12/11/07 Time: 16:34

Sample: 1 23

Included.

Included observations: 23

Variable Coefficient Std. Error t-Statistic Prob.

C 331.7015 142.5882 2.326290 0.0306

X3 1.766892 0.553402 3.192782 0.0046

X4 -1.473721 0.656624 -2.244390 0.0363

R-squared 0.684240 Mean dependent var 945.2913

Adjusted R-squared 0.652664 S.D. dependent var 224.1711

S.E. of regression 132.1157 Akaike info criterion 12.72634

Sum squared resid 349091.0 Schwarz criterion 12.87445

Log likelihood -143.3529 F-statistic 21.66965

Durbin-Watson stat 2.111635 Prob(F-statistic) 0.000010

From the results of the data, it can be seen that the adjusted coefficient of determination of the equation is the largest when X4 is introduced, and the explanatory variables have passed the test of significance, and then X1 and X2 are introduced for analysis.

X1, X3, X4

Dependent Variable: Y

Method: Least Squares

Date: 12/11/07 Time: 16:37

Sample: 1 23

Included observations: 23

Variable Coefficient Std. Error t-Statistic Prob.

C 193.6693 403.8464 0.479562 0.6370

X1 89.29944 243.6512 0.366505 0.7180

X3 1.652622 0.646003 2.558228 0.0192

X4 -1.345001 0.757634 -1.775265 0.0919

R-squared 0.686457 Mean dependent var 945.2913

Adjusted R-squared 0.636950 S.D. dependent var 224.1711

S.E. of regression 135.0712 Akaike info criterion 12.80625

Sum squared resid 346640.3 Schwarz criterion 13.00373

Log likelihood -143.2719 F-statistic 13.86591

Durbin-Watson stat 2.082104 Prob(F-statistic) 0.000050

X2, X3, X4

Dependent Variable: Y

Method: Least Squares

Date: 12/11/07 Time. 16:38

Sample: 1 23

Included observations: 23

Variable Coefficient Std. Error t-Statistic Prob.

C 62.60939 489.2088 0.127981 0.8995

X2 134.1557 232.9303 0.575948 0.5714

X3 1.886588 0.600027 3.144175 0.0053

X4 -1.596394 0.701018 -2.277251 0.0345

R-squared 0.689658 Mean dependent var 945.2913

Adjusted R-squared 0.640657 S.D. dependent var 224.1711

S.E. of regression 134.3798 Akaike info criterion 12.79599

Sum squared resid 343100.8 Schwarz criterion 12.99347

Log likelihood -143.1539 F- statistic 14.07429

Durbin-Watson stat 2.143110 Prob(F-statistic) 0.000046

From the output results, it can be seen that at the level of the explanatory variables , the p-value of the test is greater than 0.05, the explanatory variables can not pass the test of significance. Therefore it can be concluded that only two variables X3 and X4 can be introduced in the model. Then the adjusted multiple linear regression equation is:

Se= (142.5882) (0.553402) (0.656624)

T= (2.326290) (3.192782) (-2.244390)

F=21.66965 df=20

(iii). Test for Heteroskedasticity

White test for the model:

White Heteroskedasticity Test:

F-statistic 1.071659 Probability 0.399378

Obs*R-squared 4.423847 Probability 0.351673

Test Equation:

Dependent Variable: RESID^2

Method: Least Squares

Date: 12/11/07 Time. 16:53

Sample: 1 23

Included observations: 23

Variable Coefficient Std. Error t-Statistic Prob.

C 34247.50 128527.9 0.266460 0.7929

X3 247.9623 628.1924 0.394723 0.6977

X3^2 -0.071268 0.187278 -0.380548 0.7080

X4 -333.6779 714.3390 - 0.467114 0.6460

X4^2 0.121138 0.229933 0.526841 0.6047

R-squared 0.192341 Mean dependent var 15177.87

Adjusted R-squared 0.012861 S.D. dependent var 23242.54

S.E. of regression 23092.59 Akaike info criterion 23.12207

Sum squared resid 9.60E+09 Schwarz criterion 23.36892

Log likelihood -260.9038 F-statistic 1.071659

Durbin-Watson stat 1.968939 Prob(F-statistic) 0.399378

By the The test result shows that , from White's test, at the time, check the distribution table, get the critical value (20)=30.1435, because < (5)= 30.1435,so there is no heteroskedasticity in the model.

(iv). Test of autocorrelation

From the output of the model, it can be seen that the estimation results are satisfactory, both the regression equation test, and parameter significance test of the test probability, are significantly less than 0.05, D-W value of 2.111635, the level of significance = 0.05 under the check of the Durbin-Watson table, which n = 23, the number of explanatory variables for the number of 2, get Lower critical value , upper critical value , =1.543<D-W=2.111635<4 , by the DW test decision rule, the model does not have autocorrelation.

Six, analyze the model and explain the economic significance

The significance of the regression equation is: when the average real monthly income per person is unchanged, per capita disposable income per unit of increase in per capita consumption expenditure decreases by 1.473721 units; when per capita disposable income per capita is unchanged, per capita consumption expenditure per person per unit of increase in per capita disposable income per unit of increase in average real monthly income per person increases by 1.766892 units.

VII. Give targeted policy recommendations or conclusions on the issues reflected in the model

In the analysis of China's per capita consumption expenditure, it can be seen that China's economic development in the past few years has been sound, but due to a variety of reasons that lead to the current state of China's economy, there are certain problems, such as imperfect social security system leading to an irrational consumption structure; excessive savings deposits affect the influence of However, due to various reasons, there are some problems in the current situation of China's economy, such as the unreasonable consumption structure caused by the imperfect social security system, the influence of high savings deposits on residents' consumption tendency, the obstruction of consumption in the domestic market caused by the wrong investment direction and inefficiency of consumer goods industries, the constraints of conservative consumption concepts and policies, and the influence of the excessive proportion of education expenditure on residents' consumption tendency. In this regard, our country should carry out countermeasures in the following aspects of the problems in the residents' consumption

(a) Establishing and improving the social security system, and enhancing the residents' confidence in consumption

(b) Cultivating new hot spots of consumption, and expanding the residents' areas of consumption

(c) Promoting the change of the commodity consumption from self-accumulation to credit-supporting type

(d) Promoting residents' consumption at different levels

(v) Cracking the policy constraints affecting the optimization of the consumption structure

(vi) Resolving the contradiction between the lack of effective supply and the relative surplus of products