Pre-Estimation Checks

Read this page before fitting CCP. These checks focus on support and conditioning because the estimator leans on first-stage choice probabilities.

CCP can show estimation risk for reasons that are visible before optimization starts. Run these checks before treating a result as structural evidence.

Check	Why it matters
Feature rank	Reward parameters are not identified when action-dependent features are collinear.
Feature condition number	Near-collinearity inflates standard errors.
Transition row sums	Each transition row must be a valid probability distribution.
State coverage	Unvisited states require extrapolated CCPs.
State-action coverage	One-action states make counterfactual action values weakly supported.
Minimum positive CCP	Very small probabilities make the log correction unstable.
Reward normalization	Reward level and scale need a valid anchor.
Transition orientation	CCP expects transition tensors in action, state, next-state order.

Canonical Simulation Checks

The CCP results file records these pre-estimation checks. See the simulation study page for the generator script, JSON results file.

Check	Value	Status
Feature rank	4 / 4	pass
Feature condition number	4.512	pass
Transition row error	2.42e-8	pass
Observed states	21 / 21	pass
State-action coverage	1.000	pass
Action shares	0.345, 0.330, 0.325	pass
Minimum action share	0.325	pass
Minimum positive CCP	0.153	pass
Exit and absorbing anchor	true	pass

The canonical cell has full support, so the CCP evidence is about the estimator rather than a sparse first-stage policy.

Common Risk Patterns

Data with many unvisited states force CCP to extrapolate the first-stage policy. States with only one observed action make counterfactual action values weakly supported. Too little smoothing can make log corrections unstable, while too much smoothing biases the empirical policy toward uniform choice. Transition matrices with wrong orientation can produce plausible arrays and wrong economics.