This presentation is the property of its rightful owner.
Sponsored Links
1 / 69

第七讲 回归分析 PowerPoint PPT Presentation


  • 162 Views
  • Uploaded on
  • Presentation posted in: General

第七讲 回归分析. 一、线性回归分析. 线性回归是统计分析方法中最常用的方法之一。如果所研究的现象有若干个影响因素,且这些因素对现象的综合影响是线性的,则可以使用线性回归的方法建立现象 (因变量)与影响因素(自变量)之间的线性函数关系式。 由于多元线性回归的计算量比较大,所以有必要应用统计分析软件实现。. SPSS 软件中进行线性回归分析的选择项为 Analyze→Regression→Linear 。如图所示。. (一)双变量线性回归.

Download Presentation

第七讲 回归分析

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript


5468481


5468481


5468481

  • SPSSAnalyzeRegressionLinear


5468481

  • 103cmkgcm2


5468481

1

  • Y3X1X218.1


5468481

2

  • AnalyzeRegressionLinear...Linear Regression

  • yDependent

  • x1x2Indepentdent(s)

  • Method5EnterStepwiseRemoveBackwardForward

  • EnterOK


5468481

  • Statistics...

  • Plots...Y

  • Save...YY

  • Options...


5468481

3


5468481

  • X1X2Y

  • 0.94964r20.90181F=34.14499P=0.0003

  • Y=0.0687101X1+0.183756X2-2.856476


5468481

  • YYYY01Y

  • X1X2Ypre_1Y


5468481

  • YChart Carousel


5468481

  • Y 401.73967.922


5468481

  • Y 1190.017793.915


5468481


5468481

  • 1Analyze Regression LinearLinear


5468481

  • 2YDependent Independent

  • Method

    (Enter)(Remove)(Forward)(Backward)(Stepwise)

  • (Stepwise)


5468481

  • Enter()

  • Remove()

  • Forward()


5468481

  • Backward()

  • Stepwise()


5468481

  • 3StatisticsLinear Regression Statistics


5468481

  • Regression Coefficients

    Estimates ():

    Confidence intervals:95

    Covariance matrix:


5468481

Model fitRR2R2R2

R squared change:R2

Descriptives:

Part and Partial correlations

Collinearity diagnostics


5468481

Residuals

Durbin-WatsonD.W.

Casewise diagnostics: ,

Outliers outside( )standard deviations:3

All case

D.W295%


5468481

  • 4PlotsLinear RegressionPlots


5468481

  • Dependent XYYXZPRED:ZRESID:DRESID:ADJPRED:SRESIDSDRESID:


5468481

  • Standardized Residual Plots

  • Histogram:

  • Normal probability plot .

  • Produce all partial plots

  • DependentZRESID


5468481

  • 5OptionsLinear RegressionOptionsStepping Method Criteria


5468481

  • Stepping Method Criteria

  • Use probability of F:F

  • Use F value: F

  • Include constant in equation

  • Missing Values


5468481

  • 6SaveLinear RegressionSave

  • 7OK


Model summary d

Model Summary(d)

  • RR2FD.W30.993DW2.066DW


5468481

  • 33F3


5468481


5468481

  • Model

  • 1

  • 2

  • 3


5468481

  • 0.7

  • ToleranceVIF


5468481


5468481


5468481

20


5468481

1

  • 15(y,g)(x1) (x2) (x3) (x4)(x5)(ug)


5468481

2


5468481

  • SPSS


5468481

  • 19902002


5468481


5468481


5468481

1Analyze Regression Curve EstimationCurve Estimation

2SPSS


5468481

3SaveSave

4OK

Save VariablesPredicted ValuesResidualsPrediction intervals95%


5468481

Independent: X


5468481


5468481

  • 1Com

  • Display AMOVA table,


5468481

  • F


Logistic

Logistic

  • Logistic


5468481

  • Logistic


5468481

  • YYNX1X2X3X4YNX5X6


5468481

  • AnalyzeRegressionLogistic..Logistic Regression

  • yDependent;

  • x1x2x3x4x5x6Covariates


5468481

  • Method7


5468481

  • 1Enter

  • 2Forward: Conditional

  • 3Forward: LR

  • 4Forward: WaldWald

  • 5Backward: Conditional

  • 6Backward: LR

  • 7Backward: WaldWald

  • Forward: Conditional


5468481

  • OptionsLogistic Regression: Options DisplayAt last stepContinueLogistic Regression

  • OK


5468481


5468481

  • YY0X1

  • X4Y-1X1


5468481

X380.00%X693.33%22=15.276P=0.0005

Logistic

e(123.4053-30.5171X3-10.2797X6)

P =

1+ e(123.4053-30.5171X3-10.2797X6)

X33X69P=4.510-270X31X64P=0.981051


  • Login