Unit 15 Stats

Conditions for y for Regression Inference (CIs and Hypothesis Test)

L w/ equation__

• residual plot - residuals ____

I

N

• __ of ___ has no strong skewness or outliers, OR ____

E

• residual plot - residuals have consistent ____across all values of ___

R

linear with equation, scattered

independent or 10% rule

normal

• residuals, histogram, n≥30

equal standard deviation

• vertical spread, [independent variable]

random

It is considered "reasonable" to use a regression line to predict a y-value for a given x-value when the x-value falls within the ____ of the original data set.

range

Confidence Interval for Slope β

__ is found from calculator NOT the silver platter t

• df =

t*, n-2

Where is SE_b?

How is t calculated?

On bottom row of under stDev

b / SE_b

The test of H0: β = 0 also is a test of the hypothesis that ___ is ___ in the popoulation

correlation (r ), 0

PHANTOMS for Significance Test for Slope = 0

P

H

A

N

T

O

M
S

β: true slope of population regression line between [explanatory] and [response]

H0: β = 0

Ha: β ≠0, β >0, β <0

LINER

hypothesis t test for slope

T&O from silver platter

Interpret s, standard deviation of the residuals

When using the least-squares regression line to predict [response variable] from [explanatory variable], the predictions will typically be off by about [value of s] [units of response variable]

interpreting standard error of the slope, SE_b

The standard error of the slope is ___, indicating the typical difference between the sample slope and the true population slope in repeated random samples is about ___.