Obs location y

Size: px

Start display at page:

Download "Obs location y"

Ashlynn Eaton
5 years ago
Views:

1 ods rtf file='s:\webpages\~renaes\output\sas\sas kw output.rtf'; data tab331 ; input location ; cards ; ; run; 09:52 Friday, December 20, proc print ; * using npar1way to do the Kruskal-Wallis test ; proc npar1way data = tab331 wilcoxon ; class location ; var y ; exact wilcoxon / n = 2000 ; run ; * ranking the data ; proc rank data = tab331 out = rank331 ; var y ; ranks ry ; run ; * ANOVA on both the raw and ranked data ; proc anova data = rank331 ; class location ; model y ry = location ; means location / bon tukey lsd ; run ; * multiple comparisons via permutation tests ; proc multtest data = rank331 permutation pvals ; class location ; test mean(y ry) ; contrast '1 vs 2' ; contrast '1 vs 3' ; contrast '1 vs 4' ; contrast '2 vs 3' ; contrast '2 vs 4' ; contrast '3 vs 4' ; run ; ods rtf close; Obs location y

2 09:52 Friday, December 20, Obs location y

3 The NPAR1WAY Procedure 09:52 Friday, December 20, location Wilcoxon Scores (Rank Sums) for Variable y Classified by Variable location N Sum of Scores Expected Under H0 Std Dev Under H0 Mean Score Kruskal-Wallis Test Chi-Square DF 3 Pr > Chi-Square Monte Carlo Estimate for the Exact Test Pr >= Chi-Square Estimate % Lower Conf Limit % Upper Conf Limit Number of Samples 2000 Initial Seed

4 The NPAR1WAY Procedure 09:52 Friday, December 20,

5 09:52 Friday, December 20, Class Level Information Class Levels Values location Number of Observations Read 24 Number of Observations Used 24

6 09:52 Friday, December 20, Dependent Variable: y Source DF Sum of Squares Mean Square F Value Pr > F Model Error Corrected Total R-Square Coeff Var Root MSE y Mean Source DF Anova SS Mean Square F Value Pr > F location

7 09:52 Friday, December 20, Dependent Variable: ry y Rank for Variable Source DF Sum of Squares Mean Square F Value Pr > F Model Error Corrected Total R-Square Coeff Var Root MSE ry Mean Source DF Anova SS Mean Square F Value Pr > F location

8 09:52 Friday, December 20,

9 09:52 Friday, December 20, t Tests (LSD) for y Note: This test controls the Type I comparisonwise error rate, not the experimentwise error rate. Alpha 0.05 Error Degrees of Freedom 20 Error Mean Square Critical Value of t Least Significant Difference Means with the same letter are not significantly different. t Grouping Mean N location A A A

10 09:52 Friday, December 20, Tukey's Studentized Range (HSD) Test for y Note: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ. Alpha 0.05 Error Degrees of Freedom 20 Error Mean Square Critical Value of Studentized Range Minimum Significant Difference Means with the same letter are not significantly different. Tukey Grouping Mean N location A A A

11 09:52 Friday, December 20, onferroni (Dunn) t Tests for y Note: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ. Alpha 0.05 Error Degrees of Freedom 20 Error Mean Square Critical Value of t Minimum Significant Difference Means with the same letter are not significantly different. on Grouping Mean N location A A A

12 09:52 Friday, December 20,

13 09:52 Friday, December 20, t Tests (LSD) for ry Note: This test controls the Type I comparisonwise error rate, not the experimentwise error rate. Alpha 0.05 Error Degrees of Freedom 20 Error Mean Square Critical Value of t Least Significant Difference Means with the same letter are not significantly different. t Grouping Mean N location A A A

14 09:52 Friday, December 20, Tukey's Studentized Range (HSD) Test for ry Note: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ. Alpha 0.05 Error Degrees of Freedom 20 Error Mean Square Critical Value of Studentized Range Minimum Significant Difference Means with the same letter are not significantly different. Tukey Grouping Mean N location A A A

15 09:52 Friday, December 20, onferroni (Dunn) t Tests for ry Note: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ. Alpha 0.05 Error Degrees of Freedom 20 Error Mean Square Critical Value of t Minimum Significant Difference Means with the same letter are not significantly different. on Grouping Mean N location A A A

16 The Multtest Procedure 09:52 Friday, December 20, Model Information Test for continuous variables Mean t-test Degrees of Freedom Method Pooled Tails for continuous tests Two-tailed Strata weights None P-value adjustment Permutation Center continuous variables No Number of resamples Seed Contrast Coefficients location Contrast vs 2 Centered vs 3 Centered vs 4 Centered vs 3 Centered vs 4 Centered vs 4 Centered Continuous Variable Tabulations Variable location NumObs Mean Standard Deviation y y y y ry ry ry ry

17 The Multtest Procedure 09:52 Friday, December 20, p-values Variable Contrast Raw Permutation y 1 vs y 1 vs y 1 vs y 2 vs y 2 vs y 3 vs ry 1 vs ry 1 vs ry 1 vs ry 2 vs ry 2 vs ry 3 vs

proc plot; plot Mean_Illness*Dose=Dose; run;

proc plot; plot Mean_Illness*Dose=Dose; run; options pageno=min nodate formdlim='-'; Title 'Illness Related to Dose of Therapeutic Drug'; run; data Lotus; input Dose N; Do I=1 to N; Input Illness @@; output; end; cards; 0 20 101 101 101 104 104 105