Scribd
Upload
19 views16 pages

Mind Map

The document discusses causal inference methods, focusing on binary and non-binary treatment cases, and highlights the importance of avoiding selection bias and omitted variable bias in econometric analysis. It covers various estimation techniques, including Ordinary Least Squares (OLS) and Two-Stage Least Squares (TSLS), and emphasizes the significance of including control variables for accurate causal effect identification. Additionally, it addresses the bias-variance tradeoff in model complexity and the implications for statistical inference.

Uploaded by

duongdz1919
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
19 views16 pages

Mind Map

The document discusses causal inference methods, focusing on binary and non-binary treatment cases, and highlights the importance of avoiding selection bias and omitted variable bias in econometric analysis. It covers various estimation techniques, including Ordinary Least Squares (OLS) and Two-Stage Least Squares (TSLS), and emphasizes the significance of including control variables for accurate causal effect identification. Additionally, it addresses the bias-variance tradeoff in model complexity and the implications for statistical inference.

Uploaded by

duongdz1919
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Binary Treatment Case

Univariate Treatment Case


Causal Inference via
Potential Outcomes
Moment

Theoretical Linear
Regression
t Conditions
Binary Treatment

No Selectio

Non-Binary Treatment

Heterogeneo
Homogeneous Ca
on Bias Condition

2. Causal Effects

Selection Bias

ous vs.
ausal Effect

ECONOMETRICS
Orthogo

Estimation Sim
Regre

OLS Estimator

Frisch-W

Asymptotic Normality of OLS


onality Condition

mple Linear
ession (SLR)

Waugh Theorem

short= long + omitted × coefficients of


Omitted Variable Bias
omitted on included.
Exogeneity and E

The absence of conditional


selection

Adding Control Va
Important of including variable control

Omitted Variable Bias

Avoiding "Included Variable Bias"


Endogeneity

1. Ordinary Least Square


(OLS)

ariables

3. Identification Strategies for


Causal Effects
Statistical Precision of OLS Estimator
Inference

Test and confidence intervals

Interpretation
SLR

Multiple Linear
Regression (MLR)

Bilateral Simple Test

s Unilateral Simple Test

"Joint" Test

Confidence Interval

The sign of the coefficient

Its statistical significance

Th l f th ffi i t it
Condition of a valid
instrument

Instrumental Va

Identification and estimation of causal effects

Two-Stage Least Squares


(TSLS)

LATE: Local Average Treatment Effect


Causal Effects

ariables
Bias - Variance Tradeoff

Over-fitting/Under-fitting

Non-causal prediction Cross Validation

Penalized Regression
The value of the coefficient, its
practical significance

Bias: decreasing in the complexity /


richness of the model

Variance: increasing in the complexity


/ richness of the model

Over-fitting: bias term prevails

Under-fitting: variance term prevails

You might also like