Main Content

cond

Condition number of matrix

collapse all in page

Syntax

cond(A)
cond(A,P)

Description

example

cond(A) returns the 2-norm condition number of matrix A.

example

cond(A,P) returns the P-norm condition number of matrix A.

Examples

collapse all

Compute the 2-norm condition number of the inverse of the 3-by-3 magic square A.

A = inv(sym(magic(3)));
condN2 = cond(A)
condN2 =
(5*3^(1/2))/2

Use vpa to approximate the result.

vpa(condN2, 20)
ans =
4.3301270189221932338186158537647

Compute the 1-norm condition number, the Frobenius condition number, and the infinity condition number of the inverse of the 3-by-3 magic square A.

A = inv(sym(magic(3)));
condN1 = cond(A, 1)
condNf = cond(A, 'fro')
condNi = cond(A, inf)
condN1 =
16/3
 
condNf =
(285^(1/2)*391^(1/2))/60
 
condNi =
16/3

Approximate these results by using vpa.

vpa(condN1)
vpa(condNf)
vpa(condNi)
ans =
5.3333333333333333333333333333333
ans =
5.5636468855119361058627454652148
ans =
5.3333333333333333333333333333333

Hilbert matrices are examples of ill-conditioned matrices. Numerically compute the condition numbers of the 3-by-3 Hilbert matrix by using cond and vpa.

H = hilb(sym(3));
condN2 = vpa(cond(H))
condN1 = vpa(cond(H,1))
condNf = vpa(cond(H,'fro'))
condNi = vpa(cond(H,inf))
condN2 =
524.05677758606270799646154046059

condN1 =
748.0

condNf =
526.15882107972220183000899851322

condNi =
748.0

Input Arguments

collapse all

Input, specified as a number, vector, matrix, or array, or a symbolic number, variable, array, function, or expression.

One of these values 1, 2, inf, or 'fro'.

  • cond(A,1) returns the 1-norm condition number.

  • cond(A,2) or cond(A) returns the 2-norm condition number.

  • cond(A,inf) returns the infinity norm condition number.

  • cond(A,'fro') returns the Frobenius norm condition number.

More About

collapse all

Condition Number of a Matrix

Condition number of a matrix is the ratio of the largest singular value of that matrix to the smallest singular value. The P-norm condition number of the matrix A is defined as norm(A,P)*norm(inv(A),P).

Tips

  • Calling cond for a numeric matrix that is not a symbolic object invokes the MATLAB® cond function.

Version History

Introduced in R2012b

See Also

| | | |