eviews通過悉尼大學(xué)計(jì)量經(jīng)濟(jì)學(xué)作業(yè)練習(xí)手把手教會你,配套數(shù)據(jù)請到我的文庫下載課件

1、計(jì)量經(jīng)濟(jì)學(xué)作業(yè)eviews操作報(bào)告 宋泊云201118117011.首先我們打開數(shù)據(jù)2. 選擇兩家公司：panam，texaco為研究對象進(jìn)行分析3. 打開數(shù)據(jù)panam，market，rxfree4. 建立新的數(shù)據(jù)列panam-rxfree和market-rxfree5.命名新數(shù)列rirf和rmrf，對于選定的變量ri-rf和rm-rf分別用rirf和rmrf來代表6. 對于真實(shí)的capm模型7. 建立計(jì)量經(jīng)濟(jì)學(xué)方程 8. 對于選定的變量ri-rf和rm-rf當(dāng)選定panam公司的ri作為樣本值時(shí)，對于所對應(yīng)的i進(jìn)行回歸分析，結(jié)果如圖所示回答問題，當(dāng)選定公司為panam泛美航空公司時(shí)1.繪制

2、解釋變量和被解釋變量的散點(diǎn)圖2.對所選變量進(jìn)行回歸分析和t檢驗(yàn)，并報(bào)告結(jié)論Dependent Variable: RIRFMethod: Least SquaresDate: 11/15/13 Time: 19:49Sample: 1 120Included observations: 120VariableCoefficientStd. Errort-StatisticProb.  C-0.0085760.011248-0.7624610.4473RMRF0.7345080.1637394.4858460.0000R-squared0.145688  

3、  Mean dependent var-0.003322Adjusted R-squared0.138448    S.D. dependent var0.132022S.E. of regression0.122542    Akaike info criterion-1.344196Sum squared resid1.771959    Schwarz criterion-1.297738Log likelihood82.65175 &#

4、160;  Hannan-Quinn criter.-1.325329F-statistic20.12281    Durbin-Watson stat2.207557Prob(F-statistic)0.000017從中可以看出我們的解釋變量的回歸結(jié)果VariableCoefficientStd. Errort-StatisticProb.  RMRF0.7345080.1637394.4858460.0000它的t檢驗(yàn)4.485846是十分顯著的，風(fēng)險(xiǎn)系數(shù)是0.7345083. 反向回歸RMRF和RIRF，并比

5、較回歸結(jié)果結(jié)果為通過比較結(jié)果可知兩個(gè)回歸方程的VariableCoefficientStd. Errort-StatisticProb.  RIRF0.1983470.0442164.4858460.0000VariableCoefficientStd. Errort-StatisticProb.  RMRF0.7345080.1637394.4858460.00004. 周結(jié)構(gòu)穩(wěn)定測試選定顯著性水平5%，斷點(diǎn)選擇40（對應(yīng)年份1981年4月），f值檢驗(yàn)對應(yīng)水平為，所以結(jié)果還比較顯著5. 增加變量the surprise in the inflation

6、 rate, rate of growth in real oil prices and rate of growth of industrial production 也就是RPOIL, FRBIND and CPI 首先從the market_add.txt 導(dǎo)入數(shù)據(jù)注意此處的數(shù)據(jù)是60行的，用excel修改為120行數(shù)據(jù)，否則少估計(jì)很多年然后導(dǎo)入eviews6. 建立新的121行數(shù)據(jù)表并導(dǎo)入數(shù)據(jù)，計(jì)算新變量RPOIL, FRBIND and CPI 以及他們各自的增長量Growth rate in X = (Xt-Xt-1)/Xt 用新增長量創(chuàng)建變量對于變量進(jìn)行回歸回歸結(jié)果如下分析回歸結(jié)

7、果并得出結(jié)論另選一家公司重復(fù)上述操作，所以只戰(zhàn)士回歸結(jié)果，不寫過程第二題(i) If there is discrimination in loan approvals against minorities, and the appropriate factors have been controlled for, what is the sign of 1 ? (ii) Regres approve on white and report the results in the usual form. Interpret the coefficient on white. Is it stat

8、istically significant at the 1% level? Is it practically large ? Would you conclude that there is discrimination (or no discrimination) in the market for loans? Explain. VariableCoefficientStd. Errort-StatisticProb.  WHITE0.1855950.0283636.5435990.0000根據(jù)t檢驗(yàn)的結(jié)果，歧視程度在1%的水平上是顯著的，確實(shí)存在貸款市場對于白種人

9、的區(qū)別對待問題。(iii) As controls, add the variables obrat, loanprc, unem, male, married, dep, sch and cosign. Is there still evidence of discrimination against nonwhites (using a 1% level of significance)? See below, where the data is described, for the definition of these variables WHITE0.1611580.0284055.

10、6735610.0000t值檢驗(yàn)仍然顯著，多元回歸仍然存在著對非白種人的歧視和偏見。(iv)Now allow the effect of race to interact with the variable measuring other obligations as a percent of income (obrat). Is the interaction term significant at the 1% level? What do you conclude from the hypothesis test? t檢驗(yàn)效果依然顯著，顯然種族因素對于 other obligations as a percent of income (obrat).也是有影響的(v) Re-estimate the model in part (iii) with the inclusion of an interaction term between white and male. Test the null hypothesis that the coefficient on the white male interaction term is equal to zero (against a t