Valider le DML causal de la prime de risque de variance des options S&P 500
Résumé
Ce notebook spécifie et exécute une analyse d’apprentissage automatique double de l’effet de la prime de risque de variance sur les rendements des options vendues jusqu’à leur échéance. Avant l’exécution, il détermine le traitement, le résultat, les facteurs de confusion, le calendrier, le modèle de nuisance, le plan de validation croisée temporelle, l’estimateur de covariance et la procédure de réfutation par placebo. Le calcul est délégué à un exécuteur partagé qui vérifie les facteurs de confusion requis, les plis temporels, les ajustements de nuisance et l’erreur standard HAC, puis publie un artefact lié à l’identité de la requête résolue.
Un changement méthodologique central porte sur le test placebo. Le document explique que la permutation par blocs du traitement peut gonfler la variance résiduelle et rendre les estimations placebo artificiellement étroites, produisant un résultat de réfutation trop favorable. Il compare donc les statistiques t HAC, qui intègrent l’incertitude, plutôt que les estimations brutes de l’effet. Le notebook mentionne un résultat retiré pour motiver ce changement, mais son objectif est l’exécution et la validation ; l’interprétation de la nouvelle estimation causale est reportée à une analyse distincte. Les preuves sont propres à ce traitement, à ce résultat et à la population d’analyse déclarée ; la réussite de l’exécution ne suffit pas à établir une conclusion causale.
Idées clés
- La requête DML détermine son estimand, son calendrier, ses facteurs de confusion, son modèle de nuisance et son plan d’inférence avant l’ajustement.
- Les plis temporels et la covariance HAC sont définis pour tenir compte de l’horizon du résultat.
- La permutation par blocs peut fausser les estimations d’effet placebo lorsque la variance résiduelle du traitement change.
- La comparaison des statistiques t HAC du placebo et de l’effet observé intègre l’incertitude au contrôle de réfutation.
- Un artefact validé confirme que le calcul demandé a abouti, tandis que l’interprétation de fond est reportée.
Étiquettes
Texte intégral
# S&P 500 Options: Causal DML Execution
# S&P 500 Options: Causal DML Execution
This notebook estimates the effect of the variance-risk-premium treatment on the
return-to-expiry outcome. It declares the request through the shared causal boundary and exposes
the resolved estimand, timing, confounders, nuisance model, covariance design, and refutation
protocol before execution.
`11_model_analysis` interprets the causal estimates. This notebook validates the computation
and publishes its artifact only.
Prerequisites: `03_financial_features`, `04_model_based_features`, and `05_evaluation`.
```python
"""Execute the declared S&P 500 options causal DML request."""
import polars as pl
from case_studies.research import causal_supersedes
from case_studies.sp500_options.research_workflow import open_study
```
```python
EXECUTION_TIER = "canonical"
WORKSPACE: str = ""
PREVIEW_REDUCTIONS: dict = {}
# Retired by this run: the block-permutation refutation now compares the HAC t-statistic
# rather than the raw effect, so CAUSAL_RUNNER_VERSION moved and every causal identity with
# it. The rows named here hold a p-value computed on the shrunken placebo effects; this run
# supersedes them rather than correcting them, because the statistic is different, not the
# arithmetic. Read out of each registry's current canonical identity per label, 2026-09-10.
SUPERSEDES_CAUSAL: str = "d034b82943c5"
```
## Declared and resolved request
A preview must declare all sample, symbol, fold, or placebo reductions. Canonical execution uses
the complete pre-holdout analysis population.
### What `SUPERSEDES_CAUSAL` retires here
`CausalResult.one` resolves a label to exactly one canonical identity, so a refit has to name the
identity it replaces or the registry is left with two and refuses. The retired identity is
`d034b82943c5`.
What changed is the refutation statistic, not the fit. The placebo loop used to compare each
permuted run's *effect estimate* against the observed effect. Block-permuting the treatment frees
it from the controls, so the first stage can no longer predict it and its residual keeps nearly
all its variance. That residual variance is the whole denominator of the second-stage effect, so
every placebo effect is divided by a larger number than the observed one and the placebo
distribution comes out narrower than the null it stands for. The bias runs one way, toward a
refutation that reads as passed. The comparison is now on the HAC t-statistic, which carries the
denominator in it and cancels the inflation.
The retired identity fitted the same 166,105 observations and reported the same effect of 0.4098
with a HAC standard error of 0.3509, so p = 0.243 under either statistic. Its refutation p was
0.0099, the smallest value 100 draws can report, for an effect whose own t-statistic is 1.17.
Sitting at that floor is the signature. This notebook registers and hands off; the new row is
read and interpreted in `11_model_analysis`.
```python
study = open_study(execution_tier=EXECUTION_TIER, workspace=WORKSPACE or None)
request_table = pl.DataFrame(
{
"method": ["dml"],
"label": ["ret_to_expiry"],
"config_name": ["dml"],
"execution_tier": [EXECUTION_TIER],
}
)
request_table
```
```python
request = study.causal(
**request_table.row(0, named=True),
preview_reductions=PREVIEW_REDUCTIONS,
supersedes=causal_supersedes(
study,
SUPERSEDES_CAUSAL,
"ret_to_expiry",
labels=["ret_to_expiry"],
execution_tier=EXECUTION_TIER,
),
)
resolved = request.resolve()
computation = resolved.spec["computation"]
estimand = computation["estimand"]
causal_plan = pl.DataFrame(
{
"treatment": [estimand["treatment"]],
"outcome": [estimand["outcome"]],
"confounders": [", ".join(estimand["confounders"])],
"treatment_observed_at": [estimand["treatment_observed_at"]],
"outcome_horizon": [estimand["outcome_horizon"]],
"folds": [computation["cv"]["n_folds"]],
"embargo_periods": [computation["cv"]["embargo_periods"]],
"nuisance_model": [computation["model"]["class"]],
"covariance": ["HAC with the outcome horizon"],
"placebo_method": [computation["refutation"]["method"]],
"placebo_block": [computation["refutation"]["block_size"]],
"placebo_block_basis": [computation["refutation"]["block_size_basis"]],
"analysis_rows": [computation["analysis_population"]["n_rows"]],
"training_hash": [resolved.identity],
}
)
causal_plan
```
## Execute and validate
The shared DML runner fails on missing confounders, invalid temporal folds, incomplete nuisance
fits, or a non-finite HAC standard error. A cached result must match the complete resolved
identity before it can be reused.
```python
if EXECUTION_TIER == "preview" and (not WORKSPACE or not PREVIEW_REDUCTIONS):
raise ValueError("preview execution requires WORKSPACE and PREVIEW_REDUCTIONS")
result = resolved.run()
if not result.complete or result.hash != resolved.identity:
raise RuntimeError("causal execution did not publish the complete resolved request")
```
```python
artifact = pl.DataFrame(
{
"causal_hash": [result.hash],
"label": [resolved.spec["label"]],
"execution_tier": [result.execution_tier],
"complete": [result.complete],
}
)
artifact
```
The registered causal artifact is the handoff to `11_model_analysis`. No estimate or empirical
conclusion is interpreted here.Reproduit dans son intégralité avec attribution, conformément à la licence de la source. Licence: MIT
Ce résumé a été rédigé par l’agent de recherche de Stratmill à partir de la source originale ; il n’en est pas une copie.