STAT 2215Q Study Guide - Summer 2018, Comprehensive Midterm Notes - Variance, Normal Distribution, Test Cricket
STAT 2215Q
MIDTERM EXAM
STUDY GUIDE
Fall 2018
Lecture 9 – Review
CLT: under some co 1 page, 1 sided cheat sheet.
Chapter 2
Paired sample T-test: for dependent samples
2-sample T-test: for independent samples
Paired T: t-stat
t distribution:
Two-sample T:
\ /
t-stat
t distribution,
CI for
Notation:
a distribution
or
a number
Chapter 3
• What to do when violations happen
• Data transformations
• square root
• logarithm →easier to interpret
• Know assumptions very well for each test
• Know formulas
• Know deasion rules
• Chapter 4 is not on the exam
find more resources at oneclass.com
find more resources at oneclass.com
• 5 problems total
• Nor individual transformations of data sets
find more resources at oneclass.com
find more resources at oneclass.com
24
STAT 2215Q Full Course Notes
Verified Note
24 documents
Document Summary
1 page, 1 sided cheat sheet. (cid:1850)~(cid:1840)(cid:4678)(cid:1839),(cid:2870)(cid:4679) 2-sample t-test: for independent samples t distribution: = (cid:883) Paired t: (cid:2869), , t-stat = ~ (cid:3031)(cid:3032) (cid:3116) Lecture 5: independence, equal variance, normal distribution (cid:1845)(cid:1831)(cid:4666)(cid:1876) (cid:1877) (cid:4667)= (cid:1871)(cid:2869)(cid:2870)(cid:1866)(cid:2869)+(cid:1871)(cid:2870)(cid:2870)(cid:1866)(cid:2870) (cid:1845)(cid:1831)(cid:4666)(cid:1876) (cid:4667)= (cid:3117)(cid:3118)(cid:3041)(cid:3117) (cid:1845)(cid:1831)(cid:4666)(cid:1877) (cid:4667)= (cid:3118)(cid:3118)(cid:3041)(cid:3118) Welch (cid:1846) (cid:1872)(cid:1857)(cid:1871)(cid:1872) when standard deviations of samples are not the same (cid:1872)= (cid:1876) (cid:1877) . T test for equal variance (cid:1876)(cid:2869), . ,(cid:1876)(cid:3041)(cid:3117)~(cid:4666)(cid:2869),(cid:2869)(cid:2870)(cid:4667) (cid:1877)(cid:2869), . ,(cid:1877)(cid:3041)(cid:3118)~((cid:1865)(cid:2870),(cid:2870)(cid:2870)) (cid:3042): (cid:2869)(cid:2870)=(cid:2870)(cid:2870) A procedure (cid:883)5 said to be robust if its valid even when some assumptions are violated: independence. * on exam, she will specify if variance is equal or not. * on mi(cid:374)ita(cid:271) : o(cid:374) (cid:272)he(cid:272)k (cid:862) assu(cid:373)e e(cid:395)ual va(cid:396)ia(cid:374)(cid:272)e(cid:863) (cid:271)ox: normality, equal variance. In paired (cid:1872) test (correct procedure) (cid:3005) (cid:3006)(cid:4666)(cid:3005) (cid:4667)/ (cid:3041) (cid:1872)= (cid:3051) (cid:3052) . One sample (cid:1846)/ test (cid:1872)= (cid:3116) (cid:3046)/ ~(cid:1872) (cid:2869), under (cid:1834)(cid:2868) Assumptions : when to reject: rejection region, confidence interval.
