DATA2002 - WEEK4

t - test background:

1. sample from normal populations 

normal sample的sample mean也是正态分布，var 是chi-square distribution，这两个variable statistically independent

2. 

 

 3. if the population mean is   var is  and sample mean is 

  ~  

One sample test

已知population的sample mean, test sample mean与其是否有明显差异

workflow

我们通常选为0.05

 R：t.test(x , mu =  population mean  , alternative = c("two.sided", "less", "greater") )

Two sample test

检验两个sample的sample mean是否相同

workflow: 

 R：t.test(x , y , mu =  population mean  , alternative = c("two.sided", "less", "greater"),, var.equal = TRUE )

If we do not assume equal variance, var.equal = FALSE which is a Welch t-test. But this is not a propor t-statistic because of independent of the numerator.

Paired sample t-test

测试两个一一对应的样本平均值的差距

workflow:

 Confidence interval

 is false alarm rate

c = qt(1- )/2, degree of freedom)

t.test(x, mu = population mean, conf.level = 0.99)

we reject H0 if  outside the bounds

One-sided confidence intervals

用R的时候改一下alternative

The observed  signficance level (or p-value) is the value of for which the observed data is "right on the edge". The non-coverage probability (i.e. 1 - confidence interval) for which is on the boundary of the confidence interval. R example: t.test(x, mu = 375, alt = "less" , conf.level = 1-0.1589)

Rejection regions

Critical value decision rule

 

 Rejection region for test statistics

 

Rejection region for sample mean & sample scale

  和 d 拿第一个图推导即可

 power increase as n increase, false alarm rate越小，the number of se to reject H0越少

 R大法 pwr package

pwr.t.test() t-tests (one sample, 2 sample, paired)

pwr.t.test(n = NULL, d = NULL, sig.level = 0.05, power = NULL, type = c("two.sample"
,"one.sample","paired"),alternative = c("two.sided","less","greater"))

 pwr.t2n.test() t-test (two samples with unequal n)

pwr.t2n.test(n1 = NULL, n2= NULL, d = NULL, sig.level = 0.05, power = NULL, 
alternative = c("two.sided","less","greater"))

这里d是Cohen‘s d = ||/