1、.：.；貨幣供應(yīng)量對(duì)我國(guó)股票買賣金額的影響分析摘要為了研討貨幣供應(yīng)量對(duì)我國(guó)股票買賣金額的影響程度，建立經(jīng)濟(jì)模型進(jìn)展估計(jì)檢驗(yàn)，用二元回歸分析的方法，經(jīng)過OLS法和廣義差分法進(jìn)展模型修正，從而得出狹義貨幣供應(yīng)量和廣義貨幣供應(yīng)量對(duì)股票買賣額的影響變化情況。二、關(guān)鍵字：股票買賣額 狹義貨幣供應(yīng)量 廣義貨幣供應(yīng)量 三、模型建立與檢驗(yàn)由經(jīng)濟(jì)實(shí)際知，貨幣流動(dòng)性與證券市場(chǎng)親密相關(guān)，它是貨幣政策調(diào)整的重要工具，它影響著股票價(jià)錢和買賣額，貨幣供應(yīng)量越多，實(shí)踐利率下降，添加了持有貨幣的時(shí)機(jī)本錢，貨幣會(huì)由貨幣市場(chǎng)流入資本市場(chǎng)。反之，流通中的貨幣供應(yīng)量越少，貨幣會(huì)由資本市場(chǎng)流向貨幣市場(chǎng)。對(duì)我國(guó)股票買賣額建立股票買賣金額

2、與貨幣供應(yīng)量的函數(shù)關(guān)系，Y表示股票市場(chǎng)買賣額，X1表示狹義的貨幣供應(yīng)量M1，M1=流通中現(xiàn)金+企業(yè)活期存款+機(jī)關(guān)團(tuán)體部隊(duì)存款+鄉(xiāng)村存款+個(gè)人持有的信譽(yù)卡類存款；X2表示廣義貨幣供應(yīng)量，M2=M1+居民儲(chǔ)蓄存款+企業(yè)定期存款1建立模型如下表數(shù)據(jù)是1995-2021年的時(shí)間序列數(shù)據(jù)，即觀測(cè)值是延續(xù)不同年份中的數(shù)據(jù)。表一 我國(guó)股票買賣金額與貨幣供應(yīng)量資料 單位：億元年份股票買賣額狹義貨幣供應(yīng)量廣義貨幣供應(yīng)量1995812323987.160750.51996403628514.876094.919973072234826.390995.319982354438953.7104498.51999313

3、2045837.3119897.920006082753147.2134610.320013830559871.6158301.920022799070881.8185007.020033211584118.6221222.820044233495969.7254107.0200531665107278.7298755.7200690469126035.13345603.62007460556152560.1403442.22021267113166217.13475166.6資料來源：2002,2021。對(duì)時(shí)間序列數(shù)據(jù)，建立計(jì)量經(jīng)濟(jì)模型，并進(jìn)展回歸分析。我們假設(shè)先建立如下二元回歸模型： Y=

4、C+1X1+2X2+uiY股票買賣額X1狹義貨幣供應(yīng)量X2廣義貨幣供應(yīng)量Ui隨機(jī)干擾項(xiàng)根據(jù)表一中的數(shù)據(jù)，利用EVIEWS軟件，可得如表二所示結(jié)果：Dependent Variable: YMethod: Least SquaresDate: 12/06/10 Time: 15:32Sample: 1995 2021Included observations: 14VariableCoefficientStd. Errort-StatisticProb. C-115989.052021.46-2.2298090.0475X112.889798.3142611.5503230.1493X2-3.8

5、428802.956650-1.2997410.2203R-squared0.638700 Mean dependent var82079.93Adjusted R-squared0.573010 S.D. dependent var126996.2S.E. of regression82985.04 Akaike info criterion25.67812Sum squared resid7.58E+10 Schwarz criterion25.81506Log likelihood-176.7468 F-statistic9.722821Durbin-Watson stat1.00595

6、4 Prob(F-statistic)0.003701初步方程為：Y=-115989.0+12.88979x1-3.842880 x2-2.23 1.55 -1.30R2= 0.6387 F=9.7228 DW=1.0060模型檢驗(yàn)：經(jīng)濟(jì)意義檢驗(yàn)：X2的符號(hào)不符合經(jīng)濟(jì)實(shí)際的假設(shè)，因此經(jīng)濟(jì)意義檢驗(yàn)不能經(jīng)過。統(tǒng)計(jì)檢驗(yàn)：1擬合優(yōu)度檢驗(yàn)：擬合優(yōu)度R2= 0.6387，修正后的R2=0.5730，擬合效果不是很好，闡明還有其他解釋變量對(duì)被解釋變量產(chǎn)生影響。2T檢驗(yàn)：在5%的顯著程度下，臨界值t0.025(11)=2.201, x1、x2都不能經(jīng)過t檢驗(yàn)，闡明在其他解釋變量不變的情況下，廣義的貨幣供應(yīng)量和

7、狹義的貨幣供應(yīng)量對(duì)股票買賣額沒有顯著影響。3F檢驗(yàn)：在5%的顯著程度下, F0.05(2,11)=3.98,F大于臨界值，應(yīng)回絕原假設(shè)，闡明回歸方程顯著。兩個(gè)解釋變量結(jié)合起來對(duì)被解釋變量的影響是顯著的。三計(jì)量檢驗(yàn)A多重共線性檢驗(yàn)T檢驗(yàn)和F檢驗(yàn)綜合判別法，F(xiàn)檢驗(yàn)經(jīng)過，但T檢驗(yàn)不經(jīng)過，闡明模型很能夠存在著多重共線性。相關(guān)系數(shù)判別法得到相關(guān)系數(shù)矩陣如下：X1X2X110.998197X20.9981971可以看出：X1、X2之間存在嚴(yán)重的正相關(guān)。多重共線性的修正：首先對(duì)y和x1進(jìn)展回歸分析的計(jì)算結(jié)果如下，Dependent Variable: YMethod: Least SquaresDate:

8、10/29/10 Time: 22:35Sample: 1995 2021Included observations: 14VariableCoefficientStd. Errort-StatisticProb. C-84.5045948.36-1.7709990.1019X12.1028890.5131794.0977720.0015R-squared0.583214 Mean dependent var82079.93Adjusted R-squared0.548481 S.D. dependent var126996.2S.E. of regression85335.26 Akaike

9、 info criterion25.67813Sum squared resid8.74E+10 Schwarz criterion25.76942Log likelihood-177.7469 F-statistic16.79173Durbin-Watson stat1.760149 Prob(F-statistic)0.001478寫出如下回歸分析結(jié)果：Y=-84.50+2.102889X1 -1.77 4.10 R2= 0.5832 F= 16.79 DW=1.76對(duì)y和x2進(jìn)展回歸分析結(jié)果如下：Dependent Variable: YMethod: Least SquaresDate

10、: 12/06/10 Time: 16:40Sample: 1995 2021Included observations: 14VariableCoefficientStd. Errort-StatisticProb. C-71165.8845701.29-1.5571960.1454X20.7326190.1875583.9061020.0021R-squared0.559756 Mean dependent var82079.93Adjusted R-squared0.523069 S.D. dependent var126996.2S.E. of regression87703.79 A

11、kaike info criterion25.73288Sum squared resid9.23E+10 Schwarz criterion25.82418Log likelihood-178.1302 F-statistic15.25763Durbin-Watson stat1.892712 Prob(F-statistic)0.002087寫出如下回歸分析結(jié)果：Y=-71165.88+0.732619X2 -1.56 3.91 R2= 0.5598 F= 15.26 DW=1.89顯然，y與X1的R2、修正后的R2以及t值都比y與x2的好，闡明一元回歸最正確模型應(yīng)選取x1為自變量能更好的

12、闡明問題。但是它依然沒有二元回歸的模擬擬合度高。B異方差檢驗(yàn)1、圖形檢驗(yàn)法從圖可以看到，隨著X1、X2的添加，e2有添加的趨勢(shì)2、利用懷特檢驗(yàn)法，可以得到如下結(jié)果： White Heteroskedasticity Test:F-statistic35.44328 Probability0.000030Obs*R-squared13.39530 Probability0.019943Test Equation:Dependent Variable: RESID2Method: Least SquaresDate: 12/06/10 Time: 17:51Sample: 1995 2021Inc

13、luded observations: 14VariableCoefficientStd. Errort-StatisticProb. C4.27E+092.24E+091.9021780.0937X1-2988560.1338014.-2.2335790.0560X12164.9588128.25981.2861300.2344X1*X2-108.377090.89867-1.1922830.2673X21078397.500190.42.1559730.0632X2217.6572816.042071.1006850.3030R-squared0.956807 Mean dependent

14、 var5.41E+09Adjusted R-squared0.929812 S.D. dependent var6.73E+09S.E. of regression1.78E+09 Akaike info criterion45.73809Sum squared resid2.54E+19 Schwarz criterion46.01197Log likelihood-314.1666 F-statistic35.44328Durbin-Watson stat1.159629 Prob(F-statistic)0.000030可以看到，f值和卡方檢驗(yàn)的p值都小于0.05，回絕原假設(shè)的，原假設(shè)

15、是同方差，所以結(jié)果表示存在著異方差。異方差的修正：加權(quán)最小二乘法選用權(quán)數(shù) w=1/e2Dependent Variable: YMethod: Least SquaresDate: 12/06/10 Time: 18:34Sample: 1995 2021Included observations: 14Weighting series: WVariableCoefficientStd. Errort-StatisticProb. X1-8.8981401.577200-5.6417340.0002X23.7067830.5308036.9833440.0000C-15367.9610313.

16、69-1.4900550.1643Weighted StatisticsR-squared0.999242 Mean dependent var83661.60Adjusted R-squared0.999104 S.D. dependent var217324.2S.E. of regression6506.171 Akaike info criterion20.58630Sum squared resid4.66E+08 Schwarz criterion20.72324Log likelihood-141.1041 F-statistic7246.848Durbin-Watson sta

17、t1.605515 Prob(F-statistic)0.000000Unweighted StatisticsR-squared0.365105 Mean dependent var82079.93Adjusted R-squared0.249670 S.D. dependent var126996.2S.E. of regression110006.1 Sum squared resid1.33E+11Durbin-Watson stat2.119833可以看到，R2=0.9992，修正后的R2=0.9991，比原先均有所改善，并且F檢驗(yàn)和T檢驗(yàn)都經(jīng)過。修正后的模型為：y=-15367.9

18、6-8.898140 x1+3.706783x2C、自相關(guān)檢驗(yàn)1、圖示法可以看到逆轉(zhuǎn)的次數(shù)很少，很能夠存在著自相關(guān)從散點(diǎn)圖可以很明顯的發(fā)現(xiàn)，這些點(diǎn)根本都明顯分布在一三象限，所以可以得出該模型存在正的自相關(guān)。從散點(diǎn)圖可以很明顯的發(fā)現(xiàn)，這些點(diǎn)明顯分布在一三象限，所以可以得出該模型存在正的自相關(guān)。自相關(guān)的修正廣義差分法Dependent Variable: YMethod: Least SquaresDate: 12/06/10 Time: 19:07Sample(adjusted): 1996 2021Included observations: 13 after adjusting endpoi

19、ntsConvergence not achieved after 100 iterationsVariableCoefficientStd. Errort-StatisticProb. X128.540015.0449775.6571130.0003X2-7.1155062.710440-2.6252220.0276C-247880301.65E+09-0.0149900.9884AR(1)0.9973060.2093904.7629240.0010R-squared0.858288 Mean dependent var87768.92Adjusted R-squared0.811051 S.D. dependent var130311.8S.E. of regression56644