{
  "id": "hermite-polynomials",
  "title": "Hermite Polynomials",
  "lang": "en",
  "source_lang": "en",
  "summary": "Hermite polynomials are orthogonal polynomial families associated with Gaussian weights, used in probability, approximation theory, numerical integration, and quantum mechanics.",
  "category": "mathematics",
  "subcategory": "analysis",
  "type": "concept",
  "aliases": [
    "Hermite polynomial family"
  ],
  "generated_at": "2026-11-05",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/hermite-polynomials/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/hermite-polynomials/",
      "json": "https://aiwiki.page/zh/hermite-polynomials.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Hermite polynomials are two closely related families of orthogonal polynomials on the real line, distinguished by their Gaussian weight functions and normalization. The physicists’ polynomials \\(H_n(x)\\) are orthogonal with respect to \\(e^{-x^2}\\), while the probabilists’ polynomials \\(\\operatorname{He}_n(x)\\) are orthogonal with respect to \\(e^{-x^2/2}\\). They describe repeated differentiation of Gaussian functions and occur naturally in probability, numerical analysis, and quantum mechanics."
    },
    {
      "heading": "Definitions and conventions",
      "text": "For a nonnegative integer \\(n\\), the physicists’ Hermite polynomial is defined by the Rodrigues formula\n\n\\[H_n(x)=(-1)^n e^{x^2}\\frac{d^n}{dx^n}e^{-x^2}.\\]\n\nThe probabilists’ version is\n\n\\[\\operatorname{He}_n(x) =(-1)^n e^{x^2/2}\\frac{d^n}{dx^n}e^{-x^2/2}.\\]\n\nAlthough these expressions involve exponentials and derivatives, each result is a polynomial of degree \\(n\\). Their relationship is\n\n\\[H_n(x)=2^{n/2}\\operatorname{He}_n(\\sqrt{2}\\,x), \\qquad \\operatorname{He}_n(x)=2^{-n/2}H_n(x/\\sqrt{2}).\\]\n\nThe first few are:\n\n\\(n\\) | Physicists’ \\(H_n(x)\\) | Probabilists’ \\(\\operatorname{He}_n(x)\\)\n0 | \\(1\\) | \\(1\\)\n1 | \\(2x\\) | \\(x\\)\n2 | \\(4x^2-2\\) | \\(x^2-1\\)\n3 | \\(8x^3-12x\\) | \\(x^3-3x\\)\n4 | \\(16x^4-48x^2+12\\) | \\(x^4-6x^2+3\\)\n\nThus \\(H_n\\) has leading coefficient \\(2^n\\), whereas \\(\\operatorname{He}_n\\) is monic, with leading coefficient \\(1\\). Both satisfy the parity identity\n\n\\[H_n(-x)=(-1)^nH_n(x),\\]\n\nand likewise for \\(\\operatorname{He}_n\\). Even-indexed polynomials are even functions, and odd-indexed polynomials are odd.\n\nThe name honors Charles Hermite, although related polynomials appeared earlier in work by Pierre-Simon Laplace and Pafnuty Chebyshev. Because notation varies among disciplines and software libraries, the Gaussian weight or defining formula is more reliable than the symbol alone for identifying a convention."
    },
    {
      "heading": "Generating functions and explicit formulas",
      "text": "The exponential generating functions are\n\n\\[e^{2xt-t^2} =\\sum_{n=0}^{\\infty}H_n(x)\\frac{t^n}{n!}\\]\n\nand\n\n\\[e^{xt-t^2/2} =\\sum_{n=0}^{\\infty}\\operatorname{He}_n(x)\\frac{t^n}{n!}.\\]\n\nBoth series converge for all complex \\(x\\) and \\(t\\).\n\nExpanding the exponentials and collecting powers of \\(t\\) gives\n\n\\[H_n(x) =n!\\sum_{k=0}^{\\lfloor n/2\\rfloor} \\frac{(-1)^k(2x)^{n-2k}}{k!(n-2k)!},\\]\n\nand\n\n\\[\\operatorname{He}_n(x) =n!\\sum_{k=0}^{\\lfloor n/2\\rfloor} \\frac{(-1)^k x^{n-2k}}{2^k k!(n-2k)!}.\\]\n\nThese formulas make the degree and parity explicit. They also yield the values at the origin:\n\n\\[H_{2m}(0)=(-1)^m\\frac{(2m)!}{m!}, \\qquad \\operatorname{He}_{2m}(0) =(-1)^m\\frac{(2m)!}{2^m m!},\\]\n\nwhile both odd-indexed families vanish at zero."
    },
    {
      "heading": "Orthogonality and completeness",
      "text": "The physicists’ polynomials satisfy\n\n\\[\\int_{-\\infty}^{\\infty} H_m(x)H_n(x)e^{-x^2}\\,dx =\\sqrt{\\pi}\\,2^n n!\\,\\delta_{mn},\\]\n\nwhere \\(\\delta_{mn}\\) is \\(1\\) when \\(m=n\\) and \\(0\\) otherwise. The probabilists’ convention gives\n\n\\[\\int_{-\\infty}^{\\infty} \\operatorname{He}_m(x)\\operatorname{He}_n(x)e^{-x^2/2}\\,dx =\\sqrt{2\\pi}\\,n!\\,\\delta_{mn}.\\]\n\nThese identities define orthogonality under a weighted inner product. For example,\n\n\\[\\langle f,g\\rangle =\\int_{-\\infty}^{\\infty} f(x)\\overline{g(x)}e^{-x^2}\\,dx.\\]\n\nA derivation substitutes the Rodrigues formula and applies repeated integration by parts. The boundary terms vanish because Gaussian decay dominates polynomial growth.\n\nAfter normalization, the Hermite polynomials form an orthonormal basis of the corresponding weighted Hilbert space. In particular, if \\(Z\\) has the standard normal distribution, then\n\n\\[\\mathbb E[ \\operatorname{He}_m(Z)\\operatorname{He}_n(Z)] =n!\\delta_{mn}.\\]\n\nEvery measurable \\(f\\) satisfying \\(\\mathbb E[|f(Z)|^2]<\\infty\\) therefore has an expansion\n\n\\[f(x)=\\sum_{n=0}^{\\infty} a_n\\operatorname{He}_n(x), \\qquad a_n=\\frac{\\mathbb E[f(Z)\\operatorname{He}_n(Z)]}{n!},\\]\n\nwith convergence in the Gaussian-weighted mean-square norm.\n\nCompleteness does not, by itself, imply convergence at every point or uniform convergence. Those stronger conclusions require additional assumptions."
    },
    {
      "heading": "Recurrences and differential equations",
      "text": "The three-term recurrence relations are\n\n\\[H_{n+1}(x)=2xH_n(x)-2nH_{n-1}(x),\\]\n\nand\n\n\\[\\operatorname{He}_{n+1}(x) =x\\operatorname{He}_n(x)-n\\operatorname{He}_{n-1}(x).\\]\n\nTogether with the initial polynomials of degrees zero and one, these determine each family.\n\nDifferentiation lowers the degree:\n\n\\[H_n'(x)=2nH_{n-1}(x), \\qquad \\operatorname{He}_n'(x)=n\\operatorname{He}_{n-1}(x).\\]\n\nCombining these identities produces raising relations:\n\n\\[\\left(2x-\\frac{d}{dx}\\right)H_n(x)=H_{n+1}(x),\\]\n\n\\[\\left(x-\\frac{d}{dx}\\right)\\operatorname{He}_n(x) =\\operatorname{He}_{n+1}(x).\\]\n\nEach polynomial also solves a second-order ordinary differential equation:\n\n\\[H_n''(x)-2xH_n'(x)+2nH_n(x)=0,\\]\n\n\\[\\operatorname{He}_n''(x) -x\\operatorname{He}_n'(x) +n\\operatorname{He}_n(x)=0.\\]\n\nFor example, the first can be written in weighted form as\n\n\\[-\\frac{d}{dx} \\left(e^{-x^2}H_n'(x)\\right) =2n e^{-x^2}H_n(x).\\]\n\nThis connects polynomial orthogonality with the spectral properties of a differential operator."
    },
    {
      "heading": "Zeros and Gaussian quadrature",
      "text": "For \\(n\\geq1\\), every Hermite polynomial of degree \\(n\\) has \\(n\\) distinct real zeros. The zeros are symmetric about the origin, and those of consecutive degrees strictly interlace. These properties follow from orthogonality with respect to a positive weight on the real line.\n\nThe zeros of \\(H_n\\) are the nodes of Gauss–Hermite quadrature, which approximates Gaussian-weighted integrals:\n\n\\[\\int_{-\\infty}^{\\infty}e^{-x^2}f(x)\\,dx \\approx\\sum_{j=1}^{n}w_jf(x_j).\\]\n\nHere \\(x_j\\) are the zeros of \\(H_n\\), and\n\n\\[w_j= \\frac{2^{n-1}n!\\sqrt{\\pi}} {n^2[H_{n-1}(x_j)]^2}.\\]\n\nThe rule is exact whenever \\(f\\) is a polynomial of degree at most \\(2n-1\\).\n\nZeros can also be computed through a symmetric tridiagonal matrix. For the orthonormal physicists’ polynomials\n\n\\[p_k(x)=\\frac{H_k(x)}{\\sqrt{2^k k!\\sqrt{\\pi}}},\\]\n\nmultiplication by \\(x\\) satisfies\n\n\\[xp_k(x)= \\sqrt{\\frac{k+1}{2}}\\,p_{k+1}(x) +\\sqrt{\\frac{k}{2}}\\,p_{k-1}(x).\\]\n\nThe zeros of \\(H_n\\) are the eigenvalues of the \\(n\\times n\\) tridiagonal matrix with zero diagonal and off-diagonal entries\n\\(\\sqrt{1/2},\\sqrt{2/2},\\ldots,\\sqrt{(n-1)/2}\\)."
    },
    {
      "heading": "Hermite functions and the harmonic oscillator",
      "text": "Multiplying the physicists’ polynomials by a Gaussian produces the normalized Hermite functions\n\n\\[\\psi_n(x)= \\frac{H_n(x)e^{-x^2/2}} {\\pi^{1/4}\\sqrt{2^n n!}}.\\]\n\nUnlike the polynomials themselves, these functions are square-integrable with respect to ordinary Lebesgue measure. They form an orthonormal basis of \\(L^2(\\mathbb R)\\).\n\nThey satisfy\n\n\\[\\left(-\\frac{d^2}{dx^2}+x^2\\right)\\psi_n(x) =(2n+1)\\psi_n(x).\\]\n\nConsequently, they are the dimensionless stationary wavefunctions of the quantum harmonic oscillator. For mass \\(m\\), angular frequency \\(\\omega\\), and reduced Planck constant \\(\\hbar\\), the corresponding physical coordinate is scaled by\n\n\\[\\xi=\\sqrt{\\frac{m\\omega}{\\hbar}}\\,x,\\]\n\nand the energy levels are\n\n\\[E_n=\\hbar\\omega\\left(n+\\frac12\\right).\\]\n\nHermite functions are also eigenfunctions of the Fourier transform. Under the unitary convention\n\n\\[\\mathcal F[f](\\xi) =\\frac{1}{\\sqrt{2\\pi}} \\int_{-\\infty}^{\\infty}e^{-ix\\xi}f(x)\\,dx,\\]\n\nthey obey\n\n\\[\\mathcal F[\\psi_n](\\xi)=(-i)^n\\psi_n(\\xi).\\]\n\nThe eigenvalue depends on the Fourier-transform convention."
    },
    {
      "heading": "Probability and Gaussian expansions",
      "text": "The probabilists’ convention is especially convenient for Gaussian expectations. If \\(Z\\) is standard normal and \\(f\\) is sufficiently smooth, with suitable integrability and vanishing boundary terms, then\n\n\\[\\mathbb E[\\operatorname{He}_n(Z)f(Z)] =\\mathbb E[f^{(n)}(Z)].\\]\n\nFor \\(n=1\\), this reduces to the Gaussian integration-by-parts identity\n\n\\[\\mathbb E[Zf(Z)]=\\mathbb E[f'(Z)].\\]\n\nHermite polynomials also describe dependence between Gaussian variables. If \\(X\\) and \\(Y\\) are jointly standard normal with correlation \\(\\rho\\), then\n\n\\[\\mathbb E[ \\operatorname{He}_m(X)\\operatorname{He}_n(Y)] =\\delta_{mn}n!\\rho^n.\\]\n\nThis identity is useful in studying nonlinear transformations of Gaussian stochastic processes.\n\nFor independent standard normal variables \\(Z_1,\\ldots,Z_d\\), products\n\n\\[\\operatorname{He}_{\\alpha}(z) =\\prod_{j=1}^{d}\\operatorname{He}_{\\alpha_j}(z_j)\\]\n\nform a multivariate orthogonal system. They support polynomial expansions of functions of Gaussian random inputs.\n\nHermite polynomials also appear in Edgeworth and Gram–Charlier expansions, where corrections to a Gaussian probability density are expressed as a Gaussian density multiplied by polynomials. Finite truncations of these expansions need not remain nonnegative, so they do not automatically define valid probability densities."
    },
    {
      "heading": "Computation and limitations",
      "text": "Three-term recurrences are generally more useful for computation than repeated symbolic differentiation or direct evaluation of the explicit coefficient sums. Nevertheless, unnormalized Hermite polynomials grow rapidly with degree, and direct calculations can overflow or lose precision through cancellation. Normalized polynomials, scaled Hermite functions, and appropriate recurrence algorithms can improve numerical stability.\n\nThe choice of scaling is important in function approximation. Hermite functions are naturally adapted to functions that decay on an unbounded domain. A poorly matched spatial scale can make an expansion inefficient even when completeness guarantees eventual mean-square convergence.\n\nHermite polynomials should not be confused with Hermite interpolation, an interpolation method that matches both function values and derivatives. The two concepts share a historical name but are mathematically distinct."
    }
  ],
  "links": [
    "orthogonal-polynomials",
    "probability",
    "numerical-analysis",
    "quantum-mechanics",
    "derivative",
    "polynomial",
    "generating-function",
    "inner-product",
    "integration-by-parts",
    "orthonormal-basis",
    "hilbert-space",
    "normal-distribution",
    "recurrence-relation",
    "ordinary-differential-equation",
    "matrix",
    "eigenvalues-and-eigenvectors",
    "quantum-harmonic-oscillator",
    "fourier-transform",
    "expected-value",
    "correlation",
    "stochastic-process",
    "probability-density-function",
    "numerical-stability",
    "function-approximation",
    "interpolation"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}