Sas tutorial glm2
Download as pptx, pdf1 like280 views
This document discusses repeated measures analysis of variance (ANOVA). It provides an example of a 2-way repeated measures ANOVA with factors A and B. It also demonstrates how to perform a general linear model (GLM) in SAS to analyze repeated measures data, including specifying the within-subject factor in the REPEATED statement. Finally, it discusses mixed models that include both between-subject and within-subject factors and provides code for a GLM analyzing log-histamine concentrations over time in dogs receiving different drug treatments.
Sas tutorial glm2
- 1. SAS tutorial: GLM (2)
- 2. Repeated measures ? 如果每個參與者都作了某些作業／回答某 些題目。 ? 那麼這些題目／作業便稱為重複量數 （repeated measures）。 – 或受試者內（within subject）設計。 ? 可以去除來自個體差異的variance。
- 3. Example: 2-way repeated ANOVA Source SS df MS F A A / (S*A) B B / (S*B) A*B A*B / (S*A*B) S S*A Error S*B S*A*B
- 4. GLM for repeated measures data aaa; datalines; do S= 1 to 3; 5 10 20 10 do A=1 to 2; 6 11 22 13 do B=1 to 2; 4 12 24 9 input x @@; ; output; end; end; end;
- 5. GLM for repeated measures: code Between Between ods graphics on; proc glm data=AAA; proc glm data=AAA class A B S; plots=intplot; model x=A|B|S; class A B; test h=A e=A*S; model x=A|B; test h=B e=B*S; run; test h=A*B e=A*B*S; ods graphics off; run;
- 6. Result (between) : Interaction plot
- 8. Result (within) DF 被model用完了， 沒有error可以除？
- 10. Mix models ? 有些變項是受試者內設計，有些變項是受 试者间设计。
- 11. Example ? 16 dogs ? Dependent variable: – log-histamine concentration ? Independent variables: – Drug: morphine or trimethaphan (between) – Depleted: Y or N (between) – Time: measured after 0, 1, 3, or 5 min (within)
- 12. Code data dogs; Morphine Y .07 .07 .06 .07 input Drug $ Depleted $ Histamine0 Histamine1 Morphine Y .05 .07 .06 .07 Histamine3 Histamine5; Trimethaphan N .03 .62 .31 .22 LogHistamine0=log(Histamine0); Trimethaphan N .03 1.05 .73 .60 LogHistamine1=log(Histamine1); Trimethaphan N .07 .83 1.07 .80 LogHistamine3=log(Histamine3); Trimethaphan N .09 3.13 2.06 1.23 LogHistamine5=log(Histamine5); Trimethaphan Y .10 .09 .09 .08 datalines; Trimethaphan Y .08 .09 .09 .10 Morphine N .04 .20 .10 .08 Trimethaphan Y .13 .10 .12 .12 Morphine N .02 .06 .02 .02 Trimethaphan Y .06 .05 .05 .05 Morphine N .07 1.40 .48 .24 ; Morphine N .17 .57 .35 .24 Morphine Y .10 .09 .13 .14 Morphine Y .12 .11 .10 . Missing data 不用 proc glm; 先寫betwwen，再寫within class Drug Depleted; model LogHistamine0--LogHistamine5 = Drug Depleted Drug*Depleted / nouni; repeated Time 4 (0 1 3 5) polynomial / summary printe; run;
- 13. Result
- 14. Result 檢驗Y在獨變項不同水準下 的差異值變異數是否相同。