For a square matrix or a bounded linear operator A on a Banach space, the spectral radius ρ(A) is the supremum of the absolute values of points in the spectrum of A; for matrices it equals max{|λ|: λ eigenvalue of A}.

Principle

Characterized by the spectral mapping and Gelfand formula: ρ(A)=lim_{n→∞}||A^n||^{1/n} for a bounded operator and any operator norm, and it governs asymptotic growth of operator powers and resolvent behaviour near the spectrum.

Demonstration

For a diagonal matrix diag(λ1,…,λn) the spectral radius is max_i |λ_i|. A nilpotent matrix has spectral radius 0 even if its operator norm is positive. A stochastic matrix has spectral radius 1, and powers converge or cycle according to peripheral spectrum structure.

Misapplication

Equating spectral radius with operator norm or largest singular value: norms bound the spectral radius but are not equal in general; assuming ρ(A)<1 implies rapid numerical smallness of all powers without checking nonnormal behaviour and pseudospectra in infinite-dimensional contexts.

Consequence

Controls long-term behaviour of iterates A^n, stability of dynamical systems (ρ(A)<1 implies asymptotic stability for linear maps in finite dimensions), and spectral radius formula underpins spectral gap and convergence analyses.

Reversal

Considering norm growth instead of spectral radius emphasizes operator amplification in a given norm but may miss spectral-driven asymptotic rates; conversely, spectral radius zero does not force operator to be the zero operator in infinite dimensions.

Boundary

Defined for bounded operators on Banach spaces and for finite matrices; for unbounded operators spectrum and spectral radius require domain considerations. The spectral radius captures spectral magnitude but not directional amplification measured by norms or pseudospectra.

Semantic Tension

Tension with operator norm and numerical radius: operator norm measures maximal amplification in a norm, singular values reflect Euclidean amplification, while spectral radius captures asymptotic eigenvalue magnitude and may differ significantly for nonnormal operators.

Synthesis

The spectral radius is the supremal modulus of spectrum elements, computable for matrices from eigenvalues and characterized asymptotically by Gelfand’s formula; it is the primary spectral invariant controlling exponential growth rates of operator powers and stability properties.