assessing studies based on multiple regression

Question 1

what is internal validity?

Answer 1

the statistical inferences about causal effects are valid for the population being studied

Question 2

what is external validity?

Answer 2

the statistical inferences can be generalized from the population and setting studied toother populations and settings, where the “setting” refers to the legal, policy, and physical environment and related salient features.

Question 3

what does assessing threats to external validity require?

Answer 3

it requires detailed substantiated knowledge and judgement on a case by case basis

Question 4

what are the 5 threats to internal validity of regression studies?

Answer 4

ommitted variable bias
wrong functional form
errors in variables bias
sample selection bias
simultaneous causality bias

Question 5

what do all the 5 threats to internal validity imply?

Answer 5

it implies that the expected value of the error term given all values of X is not equal to 0 or that the conditional mean independence fails in which case the OLS is biased and inconsistent

Question 6

what are the 5 solutions to ommitted variable bias?

Answer 6

1) if Omitted causal variable can be measured, include it as regressor
2) if you have data on one or more controls and they are adequate (ie conditional mean independence plausibly holds), then include the control variable
3) use pananl data in which each entity is observed more than once
4) if the ommitted variable cannot be measured , use instrumental variables regression
5) run a randomised controlled experiment

Question 7

what is a bad control?

Answer 7

variables that are themselves outcome variables in the notional experiment at hand ie what would happen to your wages if you get a college degree ( the coefficient of college wont have a causal interpretation due to causal interpretation even if college degree were randomly assigned

Question 8

what is wrong functional form?

Answer 8

arises if the functional form is incorrect??

Question 9

what are the solutions to function form misspecification?

Answer 9

1) continuous dependent variables- use the appropiate non linear specificiations in X
2) discrete dependent variables: need an extension of multiple regression methods

Question 10

what is errors in variable bias?

Answer 10

economic data can often have measurement errors
ie data entry errors in administrative data, recollection errors in surveys, ambigous questions
this leads to correlation between error term and measured variable

Question 11

what are the solutions to errors in variable bais?

Answer 11

obtain better data
develop a specific model of the measurement error process. this is only possible if a lot is known about the nature of the measurement error
instrumental variables regression

Question 12

what is the missing data and sample selction bais?

Answer 12

data often missing, sometimes this data introduces bias

Question 13

what are the three cases of missing data?

Answer 13

data are missing at random
data are missing based on the value on the value of one or more X’s
data are missing based in part on the value of Y or u

Question 14

what cases of missing data dont introduce bias and why?

Answer 14

when data is missing at random or when data are missing based on the value of one or more X’s. these dont introduce bias because the standard errors are larger than they would be if the data werent missing

Question 15

what case of missing data does cause bias?

Answer 15

wehn data are missing based in part on the value of Y or u, this bias is called sample selection bias

Question 16

when does sample selection bias arise during a selection process?

Answer 16

when the selection process influences the availaiblity of data and is related to the dependent variable

Question 17

what is survivorship bias?

Answer 17

survivorship bias is a special form of sample selection bias, it is when the only ones sampled are those which have survived an event such as managed firms being measured against hold the market funds. only the surviving managed that have outperformed in the past has survived

Question 18

what are the solutions to the sample selection bias?

Answer 18

collect the sample ina way that avoids sample selection:
- obtain a true random sample, measure at the beginning of the period

Question 19

what is simultaneous causality bias?

Answer 19

if X causes Y and Y causes X too then a large ui will mean a larger Y which in turn means a larger X

Question 20

what are the solutions to simulataneous causality bias?

Answer 20

run a randomised controlled experiment- if X is chosen at random then there is no feedback from the outcome variable to Y
develop and estimate a complete model of both directions of causality- extremely difficult in practise
use instrumental variables regression to estimate the casual effect of interest

Question 21

what is the external validity requirement for a prediction model?

Answer 21

the data used to estimate the prediction model must be from the same distribution as the out of sample observation for which the prediction is made