Models and DAGs Flashcards

You may prefer our related Brainscape-certified flashcards:
1
Q

What is a Model?

A

Represents our understanding of how the world works.
- Speculations about causal relationships, correlations and sequences of events.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the elements of a Model?

A
  1. Signature
  2. Functional Relationships
  3. Probability Distributions
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What are some challenges of Models?

A
  • Uncertainty about the true causal model
  • Difficulty writing down assumptions
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is a Signature of the Model?

A

Describes variables and their ranges.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are the variables included in the Signature?

A

Exogenous and Endogenous

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is an Exogenous variable?

A

Not caused by others, can be randomly assigned (eg, treatment)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is an Endogenous variable?

A

Caused by others (eg, outcomes, covariates, mediators, moderators)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is a Functional Relationship?

A

How endogenous variables are produced (often the outcome variable).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are the approaches to identifying a functional relationship?

A
  • Structural causal models (SCMs) with DAGs
  • Potential outcomes model
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is Probability Distribution?

A

A way to describe the likelihood or chance of different outcomes occurring for a particular event or variable. It shows how probable each outcome is, often represented by a mathematical function or a table.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is a DAG?

A

Directed Acyclic Graphs represent causal relationships between variables.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are outcome variables and what is its notation?

A

(Y): variables we want to understand (dependent variables)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What are treatment variables and what is its notation?

A

(D): variables of interest that explain outcomes (independent variables)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What are moderator variables and what is its notation?

A

(X2): additional causes of the outcome unrelated to the treatment

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What are confounder variables and what is its notation?

A

(X): variables that introduce a non-causal relationship between treatment and outcome

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What are collider variables?

A

Variables caused by both treatment and outcome (conditioning on them introduces bias)

17
Q

What are mediator variables and what is its notation?

A

(M): variables along the causal path from treatment to outcome

18
Q

What are instrument variables and what is its notation?

A

(Z): exogenous variables affecting treatment, then causing outcomes (exclusion restriction)

19
Q

What is the aim of a null model?

A

Aims to reject a null hypothesis of “zero effect”

20
Q

How do you build a model?

A

Requires theory from past research or exploratory research (model building)

21
Q

Why do you build a model?

A

Informs hypotheses about relationships between treatment and outcome

22
Q

What is Reality Tracking and how do you do it?

A

Reality tracking includes using reference models that should reflect reality (m*).
Eg, Qualitative methods (interviews, focus groups) and past research inform model development.

23
Q

What are Agnostic models used for?

A

Agnostic models are used in research to test ideas, not necessarily reflect reality. They challenge existing assumptions in a field by exploring different possibilities.

24
Q

Why is it important to assess the performance of a research design?

A
  • Evaluating design performance under different models (including null models) is crucial.
  • A robust design should provide reliable evidence across various scenarios.
25
Q

What is a backdoor path?

A

A backdoor path is a hidden route in a causal relationship or an indirect way a variable (D) can influence another variable (Y) through a common cause (X). This creates a spurious correlation between D and Y, making it difficult to determine the true causal effect of D on Y.

26
Q

What is an open backdoor path?

A

An open backdoor path occurs when this indirect influence (through the confounder) is not blocked, leading to bias in estimating the true effect/

27
Q

How to close back door paths?

A
  1. Conditioning on confounders: This means taking the confounder (the variable creating the backdoor path) into account when analyzing the relationship between the other two variables.
    Colliders: Avoid controlling for colliders where it introduces bias
28
Q

What is Backdoor Criterion?

A

The backdoor criterion is a set of rules used in causal inference to identify situations where conditioning on certain variables (like confounders) will eliminate the bias caused by backdoor paths.

29
Q

What is the importance of DAGs in Causal inference?

A
  • Provides a framework for understanding causal relationships and avoiding bias.
  • Requires thorough knowledge of the data generating process and logical reasoning.
  • Understanding collider bias is crucial for accurate estimation of causal effects.
30
Q

What are the steps of the Backdoor Criterion?

A
  1. Identify paths
  2. Check for colliders
  3. Close backdoors (Conditioning on confounders or leaving colliders alone)
31
Q
A