{
  "id": "p-series",
  "title": "p-Series",
  "lang": "en",
  "source_lang": "en",
  "summary": "A p-series is an infinite series of reciprocal powers of positive integers; it converges exactly when its real exponent is greater than one.",
  "category": "mathematics",
  "subcategory": "analysis",
  "type": "concept",
  "aliases": [
    "p series",
    "reciprocal-power series"
  ],
  "generated_at": "2026-11-06",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/p-series/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/p-series/",
      "json": "https://aiwiki.page/zh/p-series.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "A p-series is an infinite series of the form\n\n\\[\\sum_{n=1}^{\\infty}\\frac{1}{n^p} =1+\\frac{1}{2^p}+\\frac{1}{3^p}+\\cdots,\\]\n\nwhere \\(p\\) is a fixed real number. Its fundamental property is that it converges if and only if \\(p>1\\). This sharp threshold makes p-series standard benchmarks for determining whether other series converge."
    },
    {
      "heading": "Definition and convergence criterion",
      "text": "The sum of a p-series is defined, when it exists, as the finite limit of its partial sums\n\n\\[S_N=\\sum_{n=1}^{N}n^{-p}.\\]\n\nEvery term is positive, so the partial sums increase with \\(N\\). They therefore either approach a finite value or increase without bound.\n\nThe convergence criterion is\n\n\\[\\sum_{n=1}^{\\infty}\\frac{1}{n^p} \\begin{cases} \\text{converges},&p>1,\\\\ \\text{diverges to }+\\infty,&p\\leq1. \\end{cases}\\]\n\nThe cases illustrate an important distinction between terms and sums:\n\n- If \\(p\\leq0\\), the terms do not approach zero, so the series cannot converge.\n- If \\(0<p\\leq1\\), the terms approach zero, but their sum still diverges.\n- If \\(p>1\\), the terms decrease sufficiently rapidly for the sum to be finite.\n\nThe case \\(p=1\\) is the harmonic series,\n\n\\[1+\\frac12+\\frac13+\\frac14+\\cdots.\\]\n\nIts divergence is a central example showing that terms approaching zero are necessary, but not sufficient, for convergence. Changing or deleting finitely many terms does not change the convergence classification."
    },
    {
      "heading": "Proofs of the convergence criterion",
      "text": "Integral comparison\n\nFor \\(p>0\\), the function \\(f(x)=x^{-p}\\) is positive, continuous, and decreasing on \\([1,\\infty)\\). The integral test therefore relates the series to the improper integral\n\n\\[\\int_1^\\infty x^{-p}\\,dx.\\]\n\nFor \\(p\\ne1\\),\n\n\\[\\int_1^M x^{-p}\\,dx =\\frac{M^{1-p}-1}{1-p}.\\]\n\nAs \\(M\\to\\infty\\), this expression approaches \\(1/(p-1)\\) when \\(p>1\\), and grows without bound when \\(0<p<1\\). For \\(p=1\\),\n\n\\[\\int_1^M\\frac{dx}{x}=\\log M,\\]\n\nwhich also grows without bound; here \\(\\log\\) denotes the natural logarithm. Together with the failure of the terms to approach zero for \\(p\\leq0\\), this proves the criterion for every real \\(p\\).\n\nGrouping terms into blocks\n\nA second proof reveals the connection with a geometric series. For \\(p>0\\), group the terms into blocks whose indices satisfy\n\n\\[2^k\\leq n<2^{k+1}.\\]\n\nEach block contains \\(2^k\\) terms, and\n\n\\[2^{-p}2^{k(1-p)} \\leq \\sum_{n=2^k}^{2^{k+1}-1}n^{-p} \\leq 2^{k(1-p)}.\\]\n\nThus the block sums are bounded above and below by constant multiples of the geometric sequence \\(2^{k(1-p)}\\). The sum over blocks converges exactly when its ratio \\(2^{1-p}\\) is less than one, equivalently when \\(p>1\\)."
    },
    {
      "heading": "Remainders and rates of growth",
      "text": "For \\(p>1\\), let\n\n\\[R_N=\\sum_{n=N+1}^{\\infty}n^{-p}\\]\n\nbe the error after summing the first \\(N\\) terms. Integral comparison gives\n\n\\[\\frac{(N+1)^{1-p}}{p-1} \\leq R_N\\leq \\frac{N^{1-p}}{p-1}.\\]\n\nConsequently,\n\n\\[R_N\\sim\\frac{N^{1-p}}{p-1},\\]\n\nwhere \\(a_N\\sim b_N\\) means \\(a_N/b_N\\to1\\).\n\nThese bounds are useful in numerical analysis. A sufficient condition for the truncation error to be at most \\(\\varepsilon>0\\) is\n\n\\[N\\geq \\left(\\frac{1}{(p-1)\\varepsilon}\\right)^{1/(p-1)}.\\]\n\nDirect summation can therefore be very slow when \\(p\\) is close to one.\n\nIntegral bounds also describe the divergent cases. For \\(0<p<1\\),\n\n\\[S_N\\sim\\frac{N^{1-p}}{1-p},\\]\n\nwhereas for \\(p=1\\),\n\n\\[S_N\\sim\\log N.\\]\n\nThe harmonic series thus diverges more slowly than any p-series with \\(0<p<1\\)."
    },
    {
      "heading": "Relationship to the Riemann zeta function",
      "text": "For real \\(p>1\\), the sum is the value of the Riemann zeta function:\n\n\\[\\zeta(p)=\\sum_{n=1}^{\\infty}n^{-p}.\\]\n\nImportant examples include\n\n\\[\\zeta(2)=\\frac{\\pi^2}{6}, \\qquad \\zeta(4)=\\frac{\\pi^4}{90}.\\]\n\nThe evaluation of \\(\\zeta(2)\\), historically known as the Basel problem, was obtained by Leonhard Euler.\n\nThe zeta function extends by analytic continuation beyond the region where its defining series converges, except for a pole at \\(1\\). Values of this continuation outside the convergence region are not ordinary sums of the corresponding divergent p-series. For example, \\(\\zeta(0)=-1/2\\) does not mean that the positive-term series \\(1+1+1+\\cdots\\) converges to \\(-1/2\\)."
    },
    {
      "heading": "Use as a comparison series",
      "text": "P-series are widely used with comparison tests:\n\n- If \\(0\\leq a_n\\leq Cn^{-p}\\) eventually, for some \\(C>0\\) and \\(p>1\\), then \\(\\sum a_n\\) converges.\n- If \\(a_n\\geq cn^{-p}\\) eventually, for some \\(c>0\\) and \\(p\\leq1\\), then \\(\\sum a_n\\) diverges.\n- If \\(a_n/n^{-p}\\to L\\), where \\(0<L<\\infty\\), then \\(\\sum a_n\\) and the p-series have the same convergence behavior.\n\nFor example,\n\n\\[\\frac{1}{n^2+3n}\\sim\\frac{1}{n^2},\\]\n\nso its series converges.\n\nAt the boundary exponent \\(p=1\\), additional factors can change the outcome. For real \\(q\\),\n\n\\[\\sum_{n=2}^{\\infty}\\frac{1}{n(\\log n)^q}\\]\n\nconverges exactly when \\(q>1\\). Thus a power exponent alone does not classify every series that resembles a p-series."
    },
    {
      "heading": "Alternating variants and applications",
      "text": "The alternating series\n\n\\[\\sum_{n=1}^{\\infty}\\frac{(-1)^{n-1}}{n^p}\\]\n\nconverges for every \\(p>0\\). It converges absolutely when \\(p>1\\) and conditionally when \\(0<p\\leq1\\). The alternating signs are essential to the extended convergence range.\n\nP-series also determine when certain discrete power-law models define a probability distribution. If a positive-integer-valued random variable \\(X\\) has probabilities\n\n\\[\\Pr(X=n)=Cn^{-p},\\]\n\na positive normalizing constant exists exactly when \\(p>1\\), in which case\n\n\\[C=\\frac{1}{\\zeta(p)}.\\]\n\nFor \\(r>0\\), its moment \\(\\mathbb{E}[X^r]\\) is finite exactly when \\(p>r+1\\), since its defining sum is proportional to\n\n\\[\\sum_{n=1}^{\\infty}n^{-(p-r)}.\\]\n\nIn particular, the expected value is finite for \\(p>2\\), and the variance is finite for \\(p>3\\)."
    }
  ],
  "links": [
    "infinite-series",
    "real-number",
    "limit",
    "harmonic-series",
    "integral",
    "natural-logarithm",
    "geometric-series",
    "numerical-analysis",
    "riemann-zeta-function",
    "leonhard-euler",
    "analytic-continuation",
    "power-law",
    "probability-distribution",
    "random-variable",
    "expected-value",
    "variance"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}