liveBook · Manning

Hypothetical condition

As explained in Causal AI

A hypothetical condition is an alternative value or circumstance imposed or considered in a counterfactual analysis.

A hypothetical condition is an alternative value or circumstance imposed or considered in a counterfactual analysis. In causal reasoning, it answers a “what if” question by specifying what would happen if a variable took a chosen value, rather than merely describing an observed association. section 8.2.3

The condition is represented as an intervention. For example, “if side-quest engagement was high” becomes an intervention setting engagement to high; the resulting hypothetical outcome concerns in-game purchases under that intervention. Ideal intervention removes the target variable’s causal influence from its parents, represented with do-notation, counterfactual subscripts, graph surgery, or a structural causal model. section 8.2.3

Distinguish the hypothetical condition from nearby parts of a causal query:

  • A factual condition records observed facts that define the target population, such as low engagement or purchases below $50. section 8.2.4
  • A hypothetical outcome states what would occur under the imposed condition, such as purchases exceeding $50. section 8.2.3
  • A counterfactual arises when the hypothetical condition or outcome conflicts with the factual conditions, such as asking about high engagement for a player observed with low engagement. section 8.2.4

In-game purchase intervention

Ask: “What would a player’s in-game purchases be if side-quest engagement were high?” The hypothetical condition is high engagement, written as an intervention on engagement; the answer describes the resulting purchase distribution. If the observed engagement is already low, the query is counterfactual because the factual and hypothetical conditions conflict. section 8.2.3 section 8.2.4

Figure 8.4 Elements of a conditional counterfactual hypothetical formalized in counterfactual notation
figure

Use the figure to inspect how factual conditions, hypothetical conditions, and hypothetical outcomes map onto counterfactual notation. section 8.2.4

How Causal AI covers it

How deep it goes
Explains definitions, intervention notation, counterfactual conflicts, and consistency, but does not develop estimation procedures.
Examples
Uses side-quest engagement and in-game purchases to contrast ordinary hypotheticals, conflicting counterfactuals, and consistent observations.
What it assumes
Assumes familiarity with variables and probability notation while defining interventions, counterfactual notation, and factual conditions.
What you can do after
Evaluate/understand causal queries by identifying their imposed condition, factual context, and hypothetical outcome.
How it connects
Connects hypothetical conditions to interventions, counterfactuals, factual conditions, hypothetical outcomes, and consistency.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Causal AI. The section links point to the parts of the book it draws on.