pith. sign in

arxiv: 2212.13255 · v1 · pith:MF2FU4OGnew · submitted 2022-12-26 · 🧮 math.NA · cs.NA

Improved Laguerre Spectral Methods with Less Round-off Errors and Better Stability

Shenghe Huang , Haijun Yu This is my paper

Pith reviewed 2026-05-24 10:11 UTC · model grok-4.3

classification 🧮 math.NA cs.NA
keywords Laguerre polynomialsspectral methodsround-off errorsnumerical stabilityelliptic equationshalf-lineorthogonal polynomialsquadrature
0
0 comments X

The pith

A modified three-term recurrence for Laguerre polynomials reduces round-off errors and supports over one thousand stable basis functions.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that altering the standard recurrence used to build generalized Laguerre polynomials and functions cuts accumulated rounding error and prevents overflow or underflow at high degrees. This change lets spectral methods employ more than one thousand Laguerre bases on an elliptic equation defined on the positive half-line while reaching accuracy near machine precision. The same modification works for Laguerre functions, and the optimal scaling parameter turns out independent of quadrature-point count in the cases where these methods converge faster than mapped Jacobi approaches.

Core claim

By introducing a modified three-term recurrence formula, the generation of generalized Laguerre polynomials and Laguerre functions becomes numerically stable, reducing round-off errors and avoiding overflow and underflow. When applied to an elliptic equation on the positive half-line, this enables the use of more than one thousand Laguerre bases with accuracy approaching machine precision. The optimal scaling factor is found to be independent of the number of quadrature points in cases where Laguerre methods show superior convergence.

What carries the argument

The modified three-term recurrence formula for generating generalized Laguerre polynomials and Laguerre functions.

If this is right

  • Elliptic equations on the half-line can be discretized with basis sizes exceeding one thousand while retaining near-machine-precision accuracy.
  • Overflow and underflow are eliminated during polynomial generation at high degrees.
  • The scaling factor that optimizes convergence does not need to be retuned when the number of quadrature points changes.
  • Laguerre spectral methods can outperform mapped Jacobi methods in convergence speed for the tested elliptic problems.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same recurrence tweak might stabilize other orthogonal polynomial families that exhibit rapid growth at high degrees.
  • Time-dependent or nonlinear problems on unbounded domains could adopt the modified basis generation to maintain stability at large basis sizes.
  • The observed independence of optimal scaling from quadrature resolution may indicate a parameter choice that remains robust across different discretizations.

Load-bearing premise

The modified recurrence preserves the orthogonality, completeness, and spectral convergence properties of the original Laguerre basis.

What would settle it

Generating one thousand Laguerre polynomials via the modified recurrence and finding that their computed orthogonality integrals deviate from exact values by more than machine epsilon would falsify the stability improvement.

read the original abstract

Laguerre polynomials are orthogonal polynomials defined on positive half line with respect to weight $e^{-x}$. They have wide applications in scientific and engineering computations. However, the exponential growth of Laguerre polynomials of high degree makes it hard to apply them to complicated systems that need to use large numbers of Laguerre bases. In this paper, we introduce modified three-term recurrence formula to reduce the round-off error and to avoid overflow and underflow issues in generating generalized Laguerre polynomials and Laguerre functions. We apply the improved Laguerre methods to solve an elliptic equation defined on the half line. More than one thousand Laguerre bases are used in this application and meanwhile accuracy close to machine precision is achieved. The optimal scaling factor of Laguerre methods are studied and found to be independent of number of quadrature points in two cases that Laguerre methods have better convergence speeds than mapped Jacobi methods.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper proposes a modified three-term recurrence for generating generalized Laguerre polynomials and associated Laguerre functions on the half-line. The modification is claimed to reduce round-off error and eliminate overflow/underflow for high degrees. The improved basis is applied to a spectral discretization of an elliptic boundary-value problem, where more than 1000 modes are reported to yield accuracy near machine precision; the optimal scaling parameter is also stated to be independent of quadrature-point count in regimes where Laguerre methods outperform mapped Jacobi methods.

Significance. If the modified recurrence is shown to be algebraically equivalent to the classical three-term relation (or to introduce only controllable perturbations that preserve weighted orthogonality), the work would enable stable, high-order spectral approximations on unbounded domains with basis cardinalities previously inaccessible due to numerical instability. The reported use of N > 1000 with near-machine-precision accuracy would constitute a concrete advance for applications requiring large Laguerre expansions.

major comments (2)
  1. [Abstract / modified-recurrence section] Abstract and the section presenting the modified recurrence: the central claim that the new recurrence 'reduces the round-off error and avoids overflow and underflow' while still generating the generalized Laguerre polynomials is load-bearing, yet no explicit formula, algebraic derivation, or forward-error analysis is supplied. Without this, it is impossible to verify that the computed basis remains orthogonal with respect to weight e^{-x} or that the spectral convergence rate for the elliptic problem is unchanged.
  2. [Elliptic-equation application] Application section (elliptic equation): the assertion that 'accuracy close to machine precision is achieved' with more than one thousand bases rests on an unexamined assumption that the modified recurrence preserves completeness and orthogonality in the weighted L2 space. No error tables, residual norms, or comparison against the classical recurrence appear in the provided description, leaving the numerical evidence for the claim unverified.
minor comments (1)
  1. [Abstract] The abstract states that the optimal scaling factor 'is independent of number of quadrature points in two cases,' but does not identify those cases or supply the supporting figures/tables.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive comments on our manuscript. We address each of the major comments below and will revise the paper accordingly to provide the requested clarifications and additional evidence.

read point-by-point responses

  1. Referee: [Abstract / modified-recurrence section] Abstract and the section presenting the modified recurrence: the central claim that the new recurrence 'reduces the round-off error and avoids overflow and underflow' while still generating the generalized Laguerre polynomials is load-bearing, yet no explicit formula, algebraic derivation, or forward-error analysis is supplied. Without this, it is impossible to verify that the computed basis remains orthogonal with respect to weight e^{-x} or that the spectral convergence rate for the elliptic problem is unchanged.

    Authors: We appreciate this observation. The manuscript introduces the modified recurrence in Section 2, but we acknowledge that a more detailed algebraic derivation and error analysis would enhance clarity. In the revised manuscript, we will include the explicit modified three-term recurrence formula, derive it step-by-step from the standard recurrence to show algebraic equivalence (with scaling factors that do not alter the generated polynomials), and provide a forward-error analysis demonstrating reduced round-off errors. This will confirm that the weighted orthogonality with respect to e^{-x} is preserved and that the spectral convergence properties for the elliptic problem remain unchanged. revision: yes

  2. Referee: [Elliptic-equation application] Application section (elliptic equation): the assertion that 'accuracy close to machine precision is achieved' with more than one thousand bases rests on an unexamined assumption that the modified recurrence preserves completeness and orthogonality in the weighted L2 space. No error tables, residual norms, or comparison against the classical recurrence appear in the provided description, leaving the numerical evidence for the claim unverified.

    Authors: We agree that strengthening the numerical evidence is beneficial. The current manuscript reports results with over 1000 bases achieving near machine precision, but to address the referee's concern, the revised version will include error tables, residual norms for the elliptic boundary-value problem, and direct comparisons between the modified and classical recurrences. Since the modified recurrence is constructed to be equivalent to the classical one, completeness and orthogonality in the weighted L2 space are preserved by design; the additional tables will verify this numerically. revision: yes

Circularity Check

0 steps flagged

No significant circularity; algorithmic modification presented as independent improvement

full rationale

The paper introduces a modified three-term recurrence as a direct change to reduce round-off error and overflow/underflow in Laguerre polynomial generation, then applies the resulting basis to an elliptic PDE achieving near-machine precision with N>1000. No equations or claims in the abstract reduce any result to a fitted parameter, self-definition, or self-citation chain. The study of optimal scaling factors is described as empirical observation across cases, without evidence that predictions are forced by construction from inputs. The central claims rest on the algebraic validity of the recurrence modification and its numerical behavior, which are not shown to collapse to the paper's own fitted values or prior self-referential results.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The abstract invokes standard properties of orthogonal polynomials and spectral methods but introduces no explicit free parameters, axioms, or invented entities beyond the scaling factor whose optimal value is reported as independent of quadrature points in two cases.

pith-pipeline@v0.9.0 · 5675 in / 1075 out tokens · 40273 ms · 2026-05-24T10:11:29.681536+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

34 extracted references · 34 canonical work pages · 1 internal anchor

  1. [1]

    Abramowitz and I

    M. Abramowitz and I. A. Stegun , Handbook of Mathematical Functions with Formulas, Graphs, and Math- ematical Tables, U.S. Department of Commerce, 1972

    work page 1972

  2. [2]

    Azaiez, J

    M. Azaiez, J. Shen, C. Xu, and Q. Zhuang , A Laguerre-Legendre spectral method for the Stokes problem in a semi-infinite channel, SIAM J. Numer. Anal., 47 (2009), pp. 271–292, https://doi.org/10.1137/070698269

    work page doi:10.1137/070698269 2009

  3. [3]

    Bao and J

    W. Bao and J. Shen , A fourth-order time-splitting Laguerre–Hermite pseudospectral method for Bose– Einstein condensates , SIAM J. Sci. Comput., 26 (2005), pp. 2010–2028, https://doi.org/10.1137/ 030601211

    work page 2005

  4. [4]

    Bao and J

    W. Bao and J. Shen , A generalized-Laguerre–Hermite pseudospectral method for computing symmetric and central vortex states in Bose–Einstein condensates , Journal of Computational Physics, 227 (2008), pp. 9778–9793, https://doi.org/10.1016/j.jcp.2008.07.017

    work page doi:10.1016/j.jcp.2008.07.017 2008

  5. [5]

    A. H. Bhrawy, D. Baleanu, and L. M. Assas , Efficient generalized Laguerre-spectral methods for solving multi-term fractional differential equations on the half line , Journal of Vibration and Control, 20 (2014), pp. 973–985, https://doi.org/10.1177/1077546313482959

    work page doi:10.1177/1077546313482959 2014

  6. [6]

    J. P. Boyd , Asymptotic coefficients of Hermite function series , Journal of Computational Physics, 54 (1984), pp. 382–410, https://doi.org/10.1016/0021-9991(84)90124-4

    work page doi:10.1016/0021-9991(84)90124-4 1984

  7. [7]

    Canuto, M

    C. Canuto, M. Y. Hussaini, A. Quarteroni, and T. A. Zang , Spectral Methods: Evolution to Complex Geometries and Applications to Fluid Dynamics , Scientific Computation, Springer-Verlag, Berlin, 2007

    work page 2007

  8. [8]

    F. Chen, J. Shen, and H. Yu , A new spectral element method for pricing european options under the Black-Scholes and Merton jump diffusion models , Journal of Scientific Computing, (2012), pp. 1–20, https://doi.org/10.1007/s10915-011-9556-5

    work page doi:10.1007/s10915-011-9556-5 2012

  9. [9]

    S. Chen, J. Shen, and L.-L. W ang , Laguerre functions and their applications to tempered fractional dif- ferential equations on infinite intervals , Journal of Scientific Computing, 74 (2018), pp. 1286–1313, https://doi.org/10.1007/s10915-017-0495-7

    work page doi:10.1007/s10915-017-0495-7 2018

  10. [10]

    Funaro , Computational aspects of pseudospectral Laguerre approximations , Applied Numerical Mathe- matics, 6 (1990), pp

    D. Funaro , Computational aspects of pseudospectral Laguerre approximations , Applied Numerical Mathe- matics, 6 (1990), pp. 447–457, https://doi.org/10.1016/0168-9274(90)90003-x

    work page doi:10.1016/0168-9274(90)90003-x 1990

  11. [11]

    Gautschi, Orthogonal polynomials: Applications and computation , Acta Numerica, 5 (1996), pp

    W. Gautschi, Orthogonal polynomials: Applications and computation , Acta Numerica, 5 (1996), pp. 45–119, https://doi.org/10.1017/S0962492900002622

    work page doi:10.1017/s0962492900002622 1996

  12. [12]

    G. H. Golub and J. H. Welsch , Calculation of Gauss quadrature rules, Math. Comp., 23 (1969), pp. 221–230, https://doi.org/10.1090/S0025-5718-69-99647-1

    work page doi:10.1090/s0025-5718-69-99647-1 1969

  13. [13]

    Gottlieb and S

    D. Gottlieb and S. A. Orszag , Numerical Analysis of Spectral Methods: Theory and Applications , SIAM, 1977

    work page 1977

  14. [14]

    Guo and Z

    B. Guo and Z. W ang , Numerical integration based on Laguerre-Gauss interpolation , Computer Methods in Applied Mechanics and Engineering, 196 (2007), pp. 3726–3741, https://doi.org/10.1016/j.cma.2006.10. 035

    work page doi:10.1016/j.cma.2006.10 2007

  15. [15]

    B. Y. Guo and H. P. Ma , Composite Legendre-Laguerre approximation in unbounded domains , J. Comput. Math., 19 (2001), pp. 101–112

    work page 2001

  16. [16]

    Guo and J

    B.-Y. Guo and J. Shen , Laguerre-Galerkin method for nonlinear partial differential equations on a semi- infinite interval, Numerische Mathematik, 86 (2000), pp. 635–654, https://doi.org/10.1007/pl00005413

    work page doi:10.1007/pl00005413 2000

  17. [17]

    Guo, Z.-Q

    B.-Y. Guo, Z.-Q. W ang, H.-J. Tian, and L.-L. W ang, Integration processes of ordinary differential equations 18 based on Laguerre-Radau interpolations, Math. Comp., 77 (2008), pp. 181–199, https://doi.org/10.1090/ S0025-5718-07-02035-2

    work page 2008

  18. [18]

    Liu, H.-y

    F.-j. Liu, H.-y. Li, and Z.-q. W ang , Spectral methods using generalized Laguerre functions for sec- ond and fourth order problems , Numer Algor, 75 (2017), pp. 1005–1040, https://doi.org/10.1007/ s11075-016-0228-2

    work page 2017

  19. [19]

    Liu, Z.-Q

    F.-J. Liu, Z.-Q. W ang, and H.-Y. Li , A fully diagonalized spectral method using generalized Laguerre functions on the half line , Adv Comput Math, 43 (2017), pp. 1227–1259, https://doi.org/10.1007/ s10444-017-9522-3

    work page 2017

  20. [20]

    H. Ma, W. Sun, and T. Tang , Hermite spectral methods with a time-dependent scaling for parabolic equa- tions in unbounded domains , SIAM J. Numer. Anal., 43 (2005), pp. 58–75, https://doi.org/10.1137/ s0036142903421278

    work page 2005

  21. [21]

    Masoumnezhad, M

    M. Masoumnezhad, M. Saeedi, H. Yu, and H. S. Nik , A Laguerre spectral method for quadratic optimal control of nonlinear systems in a semi-infinite interval , Automatika, 61 (2020), pp. 461–474, https://doi. org/10.1080/00051144.2020.1774724

    work page doi:10.1080/00051144.2020.1774724 2020

  22. [22]

    F. W. J. Olver , ed., NIST Handbook of Mathematical Functions , Cambridge University Press : NIST, Cambridge ; New York, 2010

    work page 2010

  23. [23]

    Shen , Stable and efficient spectral methods in unbounded domains using Laguerre functions , SIAM J

    J. Shen , Stable and efficient spectral methods in unbounded domains using Laguerre functions , SIAM J. Numer. Anal., 38 (2000), pp. 1113–1133, https://doi.org/10.1137/S0036142999362936

    work page doi:10.1137/s0036142999362936 2000

  24. [24]

    J. Shen, T. Tang, and L.-L. W ang , Spectral Methods: Algorithms, Analysis and Applications , no. 41 in Springer Series in Computational Mathematics, Springer, Heidelberg ; New York, 2011

    work page 2011

  25. [25]

    Shen and L.-L

    J. Shen and L.-L. W ang, Laguerre and composite Legendre-Laguerre dual-Petrov-Galerkin methods for third- order equations, Discrete & Continuous Dynamical Systems - B, 6 (2006), p. 1381, https://doi.org/10. 3934/dcdsb.2006.6.1381

    work page 2006

  26. [26]

    Shen and L.-L

    J. Shen and L.-L. W ang , Some recent advances on spectral methods for unbounded domains , Commun. Comput. Phys, 5 (2009), pp. 195–241

    work page 2009

  27. [27]

    J. Shen, Y. W ang, and H. Yu , Efficient spectral-element methods for the electronic Schr \”odinger equation, in Sparse Grids and Applications-Stuttgart 2014, Springer, 2016, pp. 265–289

    work page 2014

  28. [28]

    Szeg ˝o, Orthogonal Polynomials , vol

    G. Szeg ˝o, Orthogonal Polynomials , vol. 23 of Colloquium Publications, American Mathematical Society, Providence, Rhode Island, 4th edition ed., 1975, https://doi.org/10.1090/coll/023

    work page doi:10.1090/coll/023 1975

  29. [29]

    Tang, The Hermite spectral method for Gaussian-type functions , SIAM Journal on Scientific Computing, 14 (1993), pp

    T. Tang, The Hermite spectral method for Gaussian-type functions , SIAM Journal on Scientific Computing, 14 (1993), pp. 594–594, https://doi.org/10.1137/0914038

    work page doi:10.1137/0914038 1993

  30. [30]

    T. Tang, H. Yuan, and T. Zhou , Hermite spectral collocation methods for fractional PDEs in unbounded do- mains, Commun. Comput. Phys., 24 (2018), pp. 1143–1168, https://doi.org/10.48550/arXiv.1801.09073, https://arxiv.org/abs/1801.09073

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1801.09073 2018

  31. [31]

    W ang and B.-y

    T.-j. W ang and B.-y. Guo , Composite generalized Laguerre-Legendre pseudospectral method for Fokker- Planck equation in an infinite channel , Applied Numerical Mathematics, 58 (2008), pp. 1448–1466, https: //doi.org/10.1016/j.apnum.2007.08.007

    work page doi:10.1016/j.apnum.2007.08.007 2008

  32. [32]

    M. Xia, S. Shao, and T. Chou , Efficient scaling and moving techniques for spectral methods in unbounded domains, SIAM J. Sci. Comput., 43 (2021), pp. A3244–A3268, https://doi.org/10.1137/20M1347711

    work page doi:10.1137/20m1347711 2021

  33. [33]

    Jennings, P

    C.-l. Xu and B.-y. Guo , Mixed Laguerre-Legendre spectral method for incompressible flow in an infi- nite strip , Advances in Computational Mathematics, 16 (2002), pp. 77–96, https://doi.org/10.1023/A: 1014249613222

    work page doi:10.1023/a: 2002

  34. [34]

    Zhang, D.-q

    C. Zhang, D.-q. Gu, Z.-q. W ang, and H.-y. Li , Efficient space-time spectral methods for second-order problems on unbounded domains , J Sci Comput, 72 (2017), pp. 679–699, https://doi.org/10.1007/ s10915-017-0374-2. 19

    work page 2017