David Sichinava, Rati Shubladze
January 10, 2018
Thirteenth Meeting
Null hypothesis ⇒ Test statistic ⇒ Reference statistic ⇒ Calculating the probability of a test statistic occuring in the reference statistic
Result | Rejecting \( H_{0} \) | Retaining \( H_{0} \) |
---|---|---|
\( H_{0} \) True | Type I error | True |
\( H_{0} \) False | True | Type II error |
Little p-value, What are you trying to say, Of significance?
Stephen T. Ziliak, Roosvelt University
Neyman, J. (1937). Outline of a theory of statistical estimation based on the classical theory of probability. Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences, 236(767), 333-380.
One sample t-test
t.test(STAR$g4reading, mu = 710)
Two samples t-test
t.test(STAR$g4reading[STAR$classtype == 1],
STAR$g4reading[STAR$classtype == 2])
resume <- read.csv("resume.csv")
x <- table(resume$race, resume$call)
prop.test(x, alternative = "greater")
minwage <- read.csv("minwage.csv")
## compute proportion of full-time employment before minimum wage increase
minwage$fullPropBefore <- minwage$fullBefore /
(minwage$fullBefore + minwage$partBefore)
## same thing after minimum-wage increase
minwage$fullPropAfter <- minwage$fullAfter /
(minwage$fullAfter + minwage$partAfter)
## an indicator for NJ: 1 if it’s located in NJ and 0 if in PA
minwage$NJ <- ifelse(minwage$location == "PA", 0, 1)
## -1 Removes intercept and creates indicator variables for each category
fit.minwage <- lm(fullPropAfter ~ -1 + NJ + fullPropBefore +
wageBefore + chain, data = minwage)
## regression result
fit.minwage
summary(fit.minwage)
confint(fit.minwage)["NJ", ]