An -node Gaussian quadrature formula chooses nodes and weights for an integral with a positive weight so that every polynomial of degree at most is integrated exactly. Its nodes are the zeros of the degree- orthogonal polynomial for that weight.
For the even weight , the monic orthogonal quadratic has the form . Orthogonality to follows from oddness; orthogonality to one gives
The nodes are . The equal weights sum to , hence each is . The two-node Gaussian quadrature with quadratic weight is therefore
To verify cubic exactness, divide any cubic as with . Orthogonality removes from the integral and its nodal values vanish; linear interpolation integrates exactly. This also proves the Gaussian property rather than merely matching two moments.
Put . For every polynomial of degree at most , has degree at most and vanishes at all nodes; thus . The monic degree- orthogonal polynomial for this integral is unique: the difference of two such monic polynomials has degree below and is orthogonal to itself, hence is zero. Consequently is the monic normalization of the degree- Legendre polynomial, proving the given nodes are exactly its roots. Together the three results characterize this Gaussian quadrature rule.
Use the weighted inner product
Here the interval is nondegenerate and the indicated integrals are finite, as required for the given orthogonal polynomials. Strict positivity follows because and a nonzero polynomial cannot vanish on an interval. Since each is monic of degree , form a basis of the polynomials of degree at most . Comparing leading coefficients, expand
By orthogonality, . For , symmetry of multiplication by gives
because has degree at most and is a linear combination of . Only the last two terms remain. Their coefficients are
For the second equality, has degree at most , so its inner product with vanishes. Rearranging proves the three-term recurrence for monic orthogonal polynomials:
For this reads because and . The coefficient is therefore immaterial and is not determined by the polynomials. The usual convention is ; if a positive is desired, any positive value gives the identical recurrence. The strictly positive norm-ratio formula is the assertion for , not a division by the norm of .
With , the weighted inner product for the Chebyshev polynomials becomes
The identity and the vanishing of for positive integers give zero when , when , and when . Thus
For Gaussian quadrature, choose the distinct nodes as the zeros of the degree- orthogonal polynomial for the weight, here , and choose the weights as the weighted integrals of their Lagrange interpolation basis functions. If a polynomial has degree at most , divide it as , where both and have degree at most . Orthogonality makes the integral of zero, while its quadrature sum also vanishes; interpolation integrates exactly. This proves degree exactness. It is optimal because the squared nodal polynomial, of degree , has zero quadrature sum and strictly positive integral.
For , , so the nodes are . Their interpolation basis functions are
The weighted zeroth and second moments are and , and the first moment is zero. Integrating these three basis functions gives equal weights , hence the Chebyshev–Gauss quadrature rule
which is exact for every polynomial of degree at most .
For distinct Gaussian quadrature nodes, let be the Lagrange interpolation polynomial with . Since has degree , exactness gives
The last inequality follows from on the interval and the nonzero polynomial being nonzero on a subinterval. It does not require prior knowledge of the signs of the weights.
For , the nodes are the roots of the supplied orthogonal polynomial :
Indeed exactness for gives , and exactness for gives equality of the weights.
Linearity of the integral in each argument gives a bilinear form, and gives symmetry. Also . If a polynomial is nonzero somewhere on this interval, continuity makes strictly positive on a smaller interval, so the integral is positive. A polynomial vanishing on the whole interval is the zero polynomial. Thus the form is positive-definite and is an inner product.
Apply the Gram-Schmidt process to . Odd functions have zero integral, so is already orthogonal to , and the third orthogonal polynomial is . Their squared norms are
Therefore an orthonormal basis is
These are the first three normalized Legendre polynomials. Their distinct degrees also prove that they span the three-dimensional vector space.
For any distinct nodes, the polynomial has degree , is nonnegative, and vanishes at every node. The quadrature rule sum for is zero, but because the weight is positive and is positive except at finitely many points. Therefore no such rule is exact for all polynomials of degree at most , irrespective of the chosen weights.
For the sine weight, its first four moments are . A monic degree-two orthogonal polynomial is
since direct use of these moments gives . Its roots lie in and are
The weights follow by requiring exactness for and : and , and the symmetric nodes imply both weights are one. To prove cubic exactness, divide any cubic as , with of degree at most one. The integral of is zero by orthogonality, and its nodal values vanish; the remaining linear polynomial is integrated exactly. This is the two-node Gaussian quadrature with sine weight; the argument proves its Gaussian quadrature property without assuming a general quadrature theorem.
A sequence of real orthogonal polynomials consists of real polynomials with degree of a polynomial and
Orthogonality is relative to the specified inner product, rather than to the ordinary coefficients. A monic polynomial has leading coefficient one, so a monic orthogonal polynomial has the form . In particular, the monic degree-zero polynomial is .
Multiply the Gegenbauer differential equation by
It becomes the Sturm-Liouville form
For two polynomial solutions of degrees , multiply their equations crosswise, subtract, and integrate over . The boundary term vanishes because and . Since the eigenvalues are distinct,
Thus , , and the displayed function is the weight.
Every interior root is simple: a solution and its derivative cannot both vanish at an ordinary point of a second-order linear differential equation unless the solution is identically zero. Let be all interior roots and put . If , then has degree below , so orthogonality gives
But has no zero and constant sign on , whence the integrand has one strict sign except at finitely many points. Its integral cannot vanish, a contradiction. Therefore : all roots are real, simple, and lie in .
For fixed interpolation nodes , interpolate any as
Integrating shows that the unique weights making the quadrature rule exact through degree are
Now let the nodes be the zeros of the degree- orthogonal polynomial . Given any , polynomial division gives
Orthogonality gives , while makes the quadrature sum for vanish. The rule is already exact on , so it is exact on . This is the defining precision property of Gaussian quadrature.
The Legendre polynomial is orthogonal on to every polynomial of degree below . Suppose it changes sign at only distinct interior points , and put
choosing its overall sign so that on . This product is positive except at finitely many points, so
contradicting orthogonality because . Thus has at least sign-changing roots in ; its degree is , so these are all its roots and they are distinct.
For arbitrary distinct nodes , let
be the Lagrange basis polynomials. Every polynomial of degree below satisfies , so the required weights exist and are
They are unique because applying any such formula to recovers .
Now assume exactness for every of degree below and define . For every polynomial of degree below , the product has degree below and vanishes at every node. Therefore
There is a unique monic polynomial of degree orthogonal to all lower-degree polynomials, so is the monic normalization of and the are precisely the roots of . Taking gives
Finally has degree , so exactness gives
These are the positive weights of Gaussian quadrature.
If the monic orthogonal polynomial had fewer than distinct sign-changing zeros in , multiply the factors corresponding to those zeros to form a polynomial of degree below . Then has one sign and is not identically zero, so its positive-weight integral cannot vanish, contradicting orthogonality. Thus all zeros are distinct and lie in the open interval.
A real inner product is a bilinear symmetric form satisfying for every nonzero , and . Its norm is . An orthogonal polynomial family has nonzero members of exact degrees , with for different indices. Exact degree is part of this definition; merely belonging to the indicated polynomial vector spaces would not suffice.
Distinct degrees make linearly independent. There are of them in the -dimensional space , so they form a basis. Any lower-degree polynomial is their linear combination, and bilinearity therefore gives for every .
Use the orthogonal-projection characterization: is the best approximation to in a finite-dimensional subspace exactly when is orthogonal to . Here is already orthogonal to , so its unique best approximation is zero. For , is nonzero and therefore cannot be that best approximation. Directly,
For the specified weight, symmetry makes every odd moment vanish, while
Choose and . A monic quadratic orthogonal to has no linear term by symmetry, and orthogonality to 1 requires . Thus a valid answer is
The weight vanishes only at one point, so a nonzero polynomial has positive squared norm; the form is indeed positive definite on the polynomial vector space. Nonzero rescalings of individual family members give equally valid unnormalized answers.
With the weighted inner product , a family of orthogonal polynomials has , each nonzero, and
Continuity and positivity of the weight make the integral finite and give . Unit norm or monic leading coefficient is an additional optional normalization.
Percy Deift 2026-11-05
Percy Deift developed an exposition of nonlinear steepest descent for Riemann-Hilbert problems. His lectures explain how an analytic matrix problem with a prescribed jump across a contour can be transformed to reveal asymptotic behavior. Applications include orthogonal polynomials, integrable systems and random matrices, connecting complex analysis with large-parameter limits.[ref]
Recurrence relation 2026-10-14
A recurrence relation specifies members of a sequence in terms of previous members, together with enough initial values to determine the sequence. A three-term recurrence for orthogonal polynomials relates consecutive orders. For example, the physicists' Hermite polynomials satisfy , starting from and .
Roderick S. C. Wong Created 2026-10-13 Updated 2026-11-05
A credited contributor to chapters of the NIST Digital Library of Mathematical Functions on algebraic and analytic methods and orthogonal polynomials. These chapters connect differential equations, asymptotic methods and special-function formulas.[ref]
Rodrigues' formula Created 2026-09-24 Updated 2026-10-31
A Rodrigues formula constructs an orthogonal polynomial by differentiating a weight multiplied by a power. The Rodrigues formula for Legendre polynomials uses a derivative of . For Generalized Laguerre polynomials,