統計解析フリーソフトR入門 - nfunao.web.fc2.com

1 R . GUI R R Commander . R .. EXCEL GUI .. R .. 2. R Commander . R Commander John Fox . McMaster GUI R . R . R R . 2005 . R Commander . R Commander .. 3.. R R Commander . R . R Commander . R Commander .. R Commander . 4. R . R . R CART . ~jgb11101/files/ CRAN R . R . R Google . 5. R .. R Commander . Rcmdr .. 6. R .. 7. R .. 8. R .. ! 9. R . R CART . ~jgb11101/files/ Rconsole Rdevga . [C: Program Files R etc].. 10. R Commander . R . --internet2 --sdi . R . ("Rcmdr", contriburl= (" ")). R Commander . library(Rcmdr). 11. R Commander . [ ] . [OK] [Japan(Tsukuba)].. 12. R Commander .. 13. R Commander .. R .. 14. R Commander . R . --internet2 --sdi R_DEFAULT_PACKAGES="Rcmdr".

2 15.. R R Commander . R . R Commander . R Commander .. R Commander . 16. R Commander . R Commander .. 17. iris . Species setosa setosa setosa setosa setosa setosa setosa .. Species setosa versicolor virginica .. 18. R Commander . Graphic by (c) ( ) 19.. R .. R R . R . R Commander R . 20.. 21.. txt SPSS Minitab STATA EXCEL . Access dBase . R .. 22.. 23.. 24.. "NA" . etc . or .. 25.. EXCEL . R Commander .. 26.. EXCEL Access .. 27.. R .. 28.. 29.. 1.. datasets . 2.. iris . 3.. 30.. 31.. 32.. iris . 33.. R .. 34.. 35.. 36.. 37.. 38.. 39.. 40.. 41.. 42.. 43.. 44.. 45.. Z . 46.. 47.. 1 2 . 48.. 49.. 50.. 1 2 . 51.. 52.. > contrasts(iris$Species) #.

3 [ ] [ ]. setosa 0 0. versicolor 1 0. virginica 0 1. > contrasts(iris$Species) # . [ ] [ ]. setosa 1 0. versicolor 0 1. virginica -1 -1. > contrasts(iris$Species) # . [,1] [,2]. setosa -1 -1. versicolor 1 -1. virginica 0 2. > contrasts(iris$Species) # ..L .Q. [1,] [2,] . [3,] 53.. > summary(iris$ ) # . Min. 1st Qu. Median Mean 3rd Qu. Max. > by(iris$ , iris$Species, summary) # . INDICES: setosa Min. 1st Qu. Median Mean 3rd Qu. Max. ---------------------------------------- -------------------- INDICES: versicolor Min. 1st Qu. Median Mean 3rd Qu. Max. ---------------------------------------- -------------------- INDICES: virginica Min. 1st Qu.

4 Median Mean 3rd Qu. Max. 54.. # [ve] [vi]. > summary(lm( Species, data=iris)) setosa 0 0. versicolor 1 0. Coefficients: virginica 0 1. Estimate Std. Err t value Pr(> t ). (Intercept) <2e-16 ** se Species[ ] <2e-16 ** ve - se Species[ ] <2e-16 ** vi - se --- # [ve] [vi]. > summary(lm( Species, data=iris)) setosa 1 0. Coefficients: versicolor 0 1. virginica -1 -1. Estimate Std. Err t value Pr(> t ). (Intercept) <2e-16 ** . Species[ ] <2e-16 ** se - . Species[ ] <2e-16 ** ve - . --- Signif. codes: 0 '**' '**' '*' '.' ' ' 1. se setosa . ve versicolor vi virginica . 55.. 56.. 57.. F .. 58.. Species Min. Min. Min. Min. setosa :50. 1st Qu. 1st Qu.

5 1st Qu. 1st Qu. versicolor:50. Median Median Median Median virginica :50. Mean Mean Mean Mean 3rd Qu. 3rd Qu. 3rd Qu. 3rd Qu. Max. Max. Max. Max. 59.. mean sd 0% 25% 50% 75% 100% n 1 150. 60.. > 100*.Table/sum(.Table). setosa versicolor virginica 33 33 33. > (.Table, p=.Probs).. Chi-squared test for given probabilities data: .Table X-squared = 0, df = 2, p-value = 1. 61.. Species 0 0 0 0 0. 62.. Species setosa versicolor virginica 63.. 64.. Pearson's product-moment correlation data: iris$ and iris$ t = 43, df = 148, p-value < alternative hypothesis: true correlation is not equal to 0. 95 percent confidence interval: sample estimates: cor 65.

6 Class Age 1st 2nd 3rd Crew Child 4 4 4 4. Adult 4 4 4 4. Pearson's Chi-squared test data: .Table X-squared = 0, df = 3, p-value = 1. 66.. , , Sex = Male Class Age 1st 2nd 3rd Crew Child 2 2 2 2. Adult 2 2 2 2. , , Sex = Female Class Age 1st 2nd 3rd Crew Child 2 2 2 2. Adult 2 2 2 2. 67.. 68.. One Sample t-test data: sleep$extra t = , df = 19, p-value = alternative hypothesis: true mean is not equal to 0. 95 percent confidence interval: sample estimates: mean of x 69.. F .. 70.. 71.. Call: lm(formula = , data = iris). Residuals: Min 1Q Median 3Q Max Coefficients: Estimate Std. Error t value Pr(> t ). (Intercept) <2e-16 **. <2e-16 **. --- Signif.

7 Codes: 0 '**' '**' '*' '.' ' ' 1.. Residual standard error: on 148 degrees of freedom Multiple R-Squared: , Adjusted R-squared: F-statistic: +03 on 1 and 148 DF, p-value: <2e-16 72.. 73.. Y ~ X Y = a + bX + . Y ~ X1 + X2 Y = a + b1X1 + b2X2 + . Y ~ . Y = Y + . Y ~ X1 * X2 Y = a + b1X1 + b2X2 + b3X1X2 + . Y ~ X1 + X2 + X1*X2 . Y ~(X1 + X2)^2 . 74.. 75.. 76.. 77.. 78.. Call: glm(formula = Species, family = gaussian(identity), data = iris). Deviance Residuals: Min 1Q Median 3Q Max Coefficients: Estimate Std. Error t value Pr(> t ). (Intercept) <2e-16 **. Species[ ] <2e-16 **. Species[ ] <2e-16 **. --- Signif. codes: 0 '**' '**' '*' '.' ' ' 1.

8 (Dispersion parameter for gaussian family taken to be ). Null deviance: on 149 degrees of freedom Residual deviance: on 147 degrees of freedom AIC: Number of Fisher Scoring iterations: 2 79.. 80.. % %. (Intercept) Species[ ] Species[ ] 81.. Anova Table (Type II tests). Response: LR Chisq Df Pr(>Chisq). Species 2 < **. --- Signif. codes: 0 '**' '**' '*' '.' ' ' 1. 82.. Deviance . Analysis of Deviance Table Model 1: Species Model 2: Resid. Df Resid. Dev Df Deviance 1 147 2 148 -1 83.. Linear hypothesis test Hypothesis: -Species[ ] + Species[ ] = 0. Model 1: Species Model 2: restricted model Df Chisq Pr(>Chisq). 1 147. 2 148 -1 < **. --- Signif.

9 Codes: 0 '**' '**' '*'. '.' ' ' 1 84.. RESET .. 85.. QQ .. 86.. Residuals vs Fitted Normal Q-Q. 3. 135 135. Standardized residuals 2. 1. Residuals 0. -1. -2. 99. 99. 142. -3. 142. 1 2 3 4 5 6 -2 -1 0 1 2. Fitted values Theoretical Quantiles Scale-Location Residuals vs Leverage 142. 3. 99 135 135. Standardized residuals Standardized residuals 2. 119. 1. 0. -1. -2. 15. -3. Cook's distance 1 2 3 4 5 6 Fitted values Leverage 87. Studentized Residuals( ). -3 -2 -1 0 1 2. -2. -1. 0. t Quantiles 1. 2. QQ . 88.. Component+Residual Plot Component+Residual Plot Component+Residual( ). Component+Residual( ). Component+Residual Plot Component+Residual Plot Component+Residual( ).

10 Component+Residual( ). setosa versicolor virginica Species 89.. Added-Variable Plot Added-Variable Plot Added-Variable Plot others others others (Intercept) others others others Added-Variable Plot Added-Variable Plot Added-Variable Plot others others others others Species[ ] others Species[ ] others 90.. 2. 1. Studentized Residuals 0. -1. -2. -3. Hat-Values 91.. effect plot effect plot 5. 4. 4. 3 5 6 7 8 effect plot Species effect plot 4. 4. 3. 3 setosa versicolor virginica Species 92.