Skip to content

Saddlepoint Approximation

Definition
AI-generated

Saddlepoint approximation is a higher-order approximation technique for the tail probabilities of a statistic, built from the cumulant generating function rather than a simple normal approximation.

Topics
GWAS
Related
Case-Control Imbalance Firth Regression Minor Allele Count (MAC)

Why it matters in GWAS

Rare variants and highly imbalanced binary traits can make ordinary score-test approximations unreliable. This is why methods such as SAIGE use saddlepoint approximation to keep p-values calibrated in the extreme tails that matter for genome-wide significance.

Example usage

"Saddlepoint approximation was used to correct score-test p-values for low-frequency variants in the imbalanced case-control GWAS."

Case-Control Imbalance Firth Regression Minor Allele Count (MAC) Rare Variant Scalable and Accurate Implementation of Generalized Mixed Model (SAIGE)
Related-term network

Click any node to open that term page.

Node order. The current term is first. Next come every page linked from Related terms, in the same top-to-bottom order as that list. Then new terms appear in two-hop layers: for each of those direct neighbors (in that same order), their related terms are appended if not already listed. A third pass does the same for two-hop nodes. The walk stops when a fixed maximum number of nodes is reached.

References

  • Butler RW. (2007). Saddlepoint Approximations with Applications. Cambridge University Press.
  • GWASTutorial: Saddlepoint approximation.
← Rust Scalable and Accurate Implementation of Generalized Mixed Model (SAIGE) →

Last updated (UTC · Git history)