OLS Assumptions

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/11

encourage image

There's no tags or description

Looks like no tags are added yet.

12 Terms

1
New cards

A1 Statement

Linearity in parameters.

<p>Linearity in parameters.</p>
2
New cards

A2 Statement

Random sampling of sample size (n).

3
New cards

A3 Statement

No perfect collinearity: No Xj is an exact linear function of the others.

4
New cards

A4 Statement

Zero conditional mean.

<p>Zero conditional mean.</p>
5
New cards

A5 Statement

Homoskedasticity. It assumes that the spread, scatter, or "noise" in your regression model's residuals remains exactly the same across all values of your independent variables.

<p>Homoskedasticity. It assumes that the spread, scatter, or "noise" in your regression model's residuals remains exactly the same across all values of your independent variables.</p>
6
New cards

A6 Statement

Normality of Errors

<p>Normality of Errors</p>
7
New cards

A1 Role/Consequences if violated

Model must be linear in the betas.

8
New cards

A2 Role/Consequences if violated

A representative draw from the population.

9
New cards

A3 Role/Consequences

OLS cannot be computed if violated.

<p>OLS cannot be computed if violated.</p>
10
New cards

A4 Role/Consequences if violated

Beta hat biased = controls, better design, RCT

11
New cards

A5 Role/Consequences if violated

SEs wrong? use robust SEs (point estimate still unbiased)

12
New cards

A6 Role/Consequences if violated

Needed only for exact t/F inference in small samples.