Project OneII a: plotIIb: Survive = a + b*female + c*fare + eView menu in Equation window: representationsView menu: Coefficient tests, WaldIIc. Plot the fitted value of survive Vs. Fare: View Menu, Actual, Fitted, Residual TableSlide 7Slide 8Slide 9Confusing without labels2d. Add age to the regressionInvestigating ClassSlide 13View: actual …, select & copy fittedRegression: Survive vs. “class” & fareInvestigate Age and GenderPlot of Survive Vs. Age for FemalesRegressionSlide 19Regression: survive Vs. “class”, fare, Age of femaleBabiesSort on Babies: ProcsSlide 23InfantsSlide 25Slide 26Slide 27Slide 28Slide 29Project One Project One Fall 2008Fall 20080.00.20.40.60.81.00 200 400 600FARESURVIVETitanic Survival (one) Or Not (zero Vs. Fare in PoundsII a: plotII a: plotIIb: Survive = a + b*female + c*fare + eIIb: Survive = a + b*female + c*fare + ei. Is regression statistically significant?, yes, F-stat & probabilityii. Interpret constant term when female and fare are zero: intercept for maleiii. Is coefficient on fare statistically significant? Yes, t-stat and probabilityiv. Interpret the coefficient on fare: Probability of survival increases by 0.14 for every 100 £ v. Is the constant term significantly different from coefficient on female?Estimation Command:=====================LS SURVIVE FARE FEMALE CEstimation Equation:=====================SURVIVE = C(1)*FARE + C(2)*FEMALE + C(3)Substituted Coefficients:=====================SURVIVE = 0.001422273828*FARE + 0.5077490703*FEMALE + 0.1540123972View menu in Equation window: View menu in Equation window: representationsrepresentationsView menu: Coefficient tests, WaldView menu: Coefficient tests, WaldWald Test:Equation: UntitledNull Hypothesis: C(1)=C(2)F-statistic 447.5543 Probability 0.000000Chi-square 447.5543 Probability 0.000000IIc. Plot the fitted value of survive Vs. IIc. Plot the fitted value of survive Vs. Fare: View Menu, Actual, Fitted, Fare: View Menu, Actual, Fitted, Residual TableResidual TableSelect fare and Fitted and from Quick menu, Graph: scatter diagram0.00.51.01.50 200 400 600FAREFITTEDFitted Vs Fare from Regression:FemaleMaleConfusing without labelsConfusing without labels2d. Add age to the regression2d. Add age to the regression2e: is it women and children first?Investigating ClassInvestigating Class•Class variable: 1, 2, 3•Genr first = 1*(class=1) + 0*(class>1)•Genr third = 1*(class=3) + 0*(class<3)•Genr second = 1-first-secondView: actual …, select & copy fitted View: actual …, select & copy fittedRegression: Survive vs. “class” & fareRegression: Survive vs. “class” & fare0.20.40.60.81.01.20 200 400 600FAREFITTEDCLASSfirstsecondthirdInvestigate Age and GenderInvestigate Age and Gender•Agefem=age*femalePlot of Survive Vs. Age for FemalesPlot of Survive Vs. Age for FemalesRegressionRegression0.00.51.01.50 20 40 60 80AGEFEMFITAGEFEMfirstsecondthirdRegression: survive Vs. “class”, fare, Regression: survive Vs. “class”, fare, Age of femaleAge of femaleBabiesBabiesSort on Babies: ProcsSort on Babies: ProcsInfantsInfants0204060801001200 10 20 30 40 50 60 70 80Series: AGESample 1 1046Observations 1046Mean 29.88113Median 28.00000Maximum 80.00000Minimum 0.166700Std. Dev. 14.41350Skewness 0.407087Kurtosis 3.140518Jarque-Bera 29.75106Probability 0.000000Histogram of age0.20.40.60.81.01.21.40 50 100 150 200FAREFITBABYFirstsecondThirdRegression of Survive Vs. Fare and "class" For
View Full Document