STA221 Exam Review Sheet

2 Pages
320 Views
Unlock Document

Department
Statistics
Course
STA221H5
Professor
Olga Fraser
Semester
Winter

Description
For CRD, indep random samples need to be selected for each treatment OR treatments have to be randomly assigned to experimental unitstiog H =Type II error; Multiple comparisons: controls for TypeI error called "experimentwise error rate" ot CI interpretation for (µ -µ1):21) -ve/+ve =cant conclude that either mean is sig. larger, comparisonwise both µ are not sig. diff [if centre of interval <0 then bigger µ value is larger, if >0 then smaller µ value is larger; 2) both -ve= µ is 2ig larger than µ 3) bo1h +ve= µ is sig. l1rger than µ 2 Factors: variables whose effect on the response is of interest to the experimenter; Qualitative factors: numerical; Factor levels: values of factor utilized in the experiment Not satisfied=nonparametric (Kruskal-Wallis H-Test (levels of qualitative factors=non-numerical e.g. levels of qualitative factor Location are N/E/W/S)(levels of quantitative factors are numerical e.g. #of…, the GPAs); Levels of Factor: specific values of the factor; Treatments: factor-level combinations(if one factor, then levels=treatments) Designed: control over levels of factor Obs: no control Completely Randomized Design: design where an independent random sample is selected for each level of the factor Model in CRD: y = µij ϵ (iesidijl: ) SST (sumofsquaresfortreatments): calculates variation between treatment means SST= SSE (sumofsquaresforerror): measures sampling variability within the treatment due to sampling errors SSE=(n -1)1 12 + (n2-1)s2+…(n -1ks k2 [sk2sample variances for the k treatments) MST(meansquarefortreatments): measures variability among the treatment means MST=SST/k-1 MSE(meansquareforerror): measures sampling variability within the treatments MSE=SSE/n-k ~ [SSE=SS (total)T] F=MST/MSE ANOVA=robustmethod when assumption of normality isnt satisfied, Boxplots=check for equal variances Number of pairwise comparisons of treatment means: c= k(k-1)/2 Randomized Block Design: Assumptions: Blocks are rando
More Less

Related notes for STA221H5

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit