Exam Prep

Question 1

Answer 1

Question 2

When to use z, when to use t

Answer 2

n is considered large from 30 and on

Question 3

LSE of β_0? How
To obtain

Answer 3

From method of least squares

Question 4

LSE of β_1? How to obtain?

Answer 4

From method of least squares

Question 5

Difference between distributions of β^_0 and μ^_0 for SLRM

Answer 5

In β^_0 var is same except Xbar squared over S_xx
(Not x_0 minus xBar squared)

Question 6

a = ?

Answer 6

Question 7

Categorical rvs in a linear model have what effect on β^- (vector)

Answer 7

Result in 2 β values for each categorical

Question 8

Plotting a linear model w a categorical variable
E(Y) against and any continuous variable

Answer 8

Doubles the number of parallel lines (assuming no interaction)

Question 9

Given usual y=xβ + ε fit, where ε takes usual normal dist
Sketch of residual plot against true model is

Answer 9

Plotting residuals against fitted values shows variance decreasing as fitted values increase
Non constant variation

Question 10

Given usual y=xβ + ε fit, where ε takes usual normal dist
Sketch of residual plot against true error distribution is t dist w 3 degrees of freedom

Answer 10

Normal plot of residuals shows an inverted s shape

Question 11

Given usual y=xβ + ε fit, where ε takes usual normal dist
Sketch of residual plot against true linear predictor should have an additional term β₁₁x²₁

Answer 11

Plot residual v x_1 shows a U or inverted U