22장 검정력(Power)과 표본수(Sample Size)

[안재형]

### Paired T-Test

> power.t.test(delta=2, sd=sqrt(2)*1.2, power=0.8)

     Two-sample t test power calculation

              n = 12.34069
          delta = 2
             sd = 1.697056
      sig.level = 0.05
          power = 0.8
    alternative = two.sided

 NOTE: n is number in *each* group

> power.t.test(delta=2,sd=sqrt(2)*1.2,power=0.9)

     Two-sample t test power calculation

              n = 16.15444
          delta = 2
             sd = 1.697056
      sig.level = 0.05
          power = 0.9
    alternative = two.sided

 NOTE: n is number in *each* group

> power.t.test(delta=2,sd=sqrt(2)*1.2,power=0.9,sig.level=0.01)

     Two-sample t test power calculation

              n = 23.14624
          delta = 2
             sd = 1.697056
      sig.level = 0.01
          power = 0.9
    alternative = two.sided

 NOTE: n is number in *each* group

> power.t.test(delta=2,sd=sqrt(2)*1.2,power=0.9,alter="one.sided")

Two-sample t test power calculation

n = 13.06832

delta = 2

sd = 1.697056

sig.level = 0.05

power = 0.9

alternative = one.sided

NOTE: n is number in *each* group

> power.t.test(n=13,delta=2,sd=sqrt(2)*1.2,alt="one.sided")

     Two-sample t test power calculation

              n = 13
          delta = 2
             sd = 1.697056
      sig.level = 0.05
          power = 0.8985622
    alternative = one.sided

 NOTE: n is number in *each* group

### Two-Sample T-Test

> power.t.test(delta=2,sd=1.2,power=0.8)

     Two-sample t test power calculation

              n = 6.76095
          delta = 2
             sd = 1.2
      sig.level = 0.05
          power = 0.8
    alternative = two.sided

 NOTE: n is number in *each* group

### One-Way ANOVA

> power.anova.test(groups=3, between.var=2, within.var=1.5, power=.8)

     Balanced one-way analysis of variance power calculation

         groups = 3
              n = 4.767866
    between.var = 2
     within.var = 1.5
      sig.level = 0.05
          power = 0.8

 NOTE: n is number in each group

# Between/Within-group variance 

> out = lm(weight ~ group, data=PlantGrowth)
> anova(out)
Analysis of Variance Table

Response: weight
               Df  Sum Sq Mean Sq F value  Pr(>F) 
group        2  3.7663    1.8832      4.8461   0.01591 *
Residuals 27 10.4921   0.3886                 

### 두 비율의 차이 검정

> power.prop.test(p1=0.25, p2=0.5, power=0.7)

     Two-sample comparison of proportions power calculation

              n = 45.63026
             p1 = 0.25
             p2 = 0.5
      sig.level = 0.05
          power = 0.7
    alternative = two.sided

 NOTE: n is number in *each* group

[강성찬]

잘 보고 갑니다.^^

[안재형]

설명은 책에 있어서 코드와 결과물만 올렸습니다^^ 근데 R에서 기본적으로 제공하는 함수로는 sample size계산 방법은 그리 다양하지 않더군요.

[이정은]

Fisher's exact test 예시는 없나요?