The stable trace formula is the endoscopic reorganization of the invariant into distributions that depend only on stable conjugacy and stable packet data.

Schematically,

IG(f)=eι(G,e)S^He(fe),I^G(f) = \sum_{\mathfrak e} \iota(G,\mathfrak e)\, \widehat S^{H_{\mathfrak e}}(f^{\mathfrak e}),

where e\mathfrak e ranges over elliptic , fef^{\mathfrak e} is an of ff, and S^He\widehat S^{H_{\mathfrak e}} is a for the endoscopic group. The actual formula includes , central, and measure data suppressed here.

Why stabilization is needed

The invariant trace formula is invariant under conjugating a , but its individual geometric terms can distinguish rational conjugacy classes inside one . On the spectral side, individual characters inside an LL-packet are likewise unstable.

Fourier analysis on the separates these unstable terms into . Endoscopic transfer identifies each piece with a stable distribution on an endoscopic group.

Inputs

Stabilization uses:

Status and scope

Arthur developed the stabilization of the general invariant trace formula, initially conditional on fundamental lemmas that are now theorems in the required standard settings. Specialized stable formulas may impose , , test-function, or characteristic hypotheses. “The stable trace formula” names this framework and its precise instances, not one hypothesis-free finite sum.

Consequences

Stable comparison makes packet characters visible and underlies Arthur's classification for symplectic and , Mok's classification for quasi-split unitary groups, and many cases of automorphic transfer.

References
  1. James Arthur, “A stable trace formula III: proof of the main theorems,” Annals of Mathematics 158 (2003), 769–873. Clay copy.
  2. James Arthur, “An introduction to the trace formula,” §§27–29. Clay Mathematics Proceedings.