4. Experiments then Standard Errors

Question 1

Diff between unbiasedness and LLN

Answer 1

Unbiasedness means that when samples of any size are used to create an estimate, the average of those sample estimates is the population parameter.

LLN applies to a single sample, large sample size implies the sample estimate approaches the relevant population parameter

Question 2

Answer 2

Question 3

Answer 3

Question 4

Random sampling and random assignment difference

Answer 4

Sampling just decides who enters the dataset, whereas random assignment decides who is treated (external validity)

Random sampling creates representative data, while random assignment removes bias by brekaing relationship between treatment and unobserved variables (internal validity)

Question 5

What is sampling randomness

Answer 5

Variability in estimates caused by using different samples from the same population

Want to know if this difference in meaningless

Question 6

Variance and s.d. formulae

Answer 6

Question 7

Answer 7

Question 8

Answer 8

Question 9

What is the standard error?

Answer 9

Estiamte of the standard deviation (x bar)

SE(x bar) = sample estimate of standard deviation/ square root n

Question 10

What is the variance of the difference in sample averages for treated and control groups? Express the formula

Answer 10

Question 11

What is the standard deviation of the difference in sample average for treated and control groups

Answer 11

Question 12

What is the standard error of the difference in sample average for treated and control groups? Express the formulae fully

Answer 12

PLugging in sample estimate for both standard deviations^2

selection bias equation

Question 13

Question 14

Answer 14

Question 15

Answer 15

Question 16

Answer 16