Finite-resolution information about a common clock: alias depth, holding-time loss, and the sampled Ehrenfest chain

A mix of human-written and AI-generated textHuman understanding: all partsmath.ST — Statistics Theorymath.PR — Probability

Contributed by Samuil Petkov ↗

Version 1 / Nov 19, 2026 / CC BY 4.0

Abstract

We study how much Fisher information about a common rate multiplier survives when a reaction system is observed only at discrete times. For a finite known mark law, the first loss at short sampling times is governed by an alias depth hh, the smallest length mismatch among reaction-count fibres: the unrecovered fraction of reaction-count information is 1−Fp(λ)=Cpλh−1+O(λh)1-F_p(\lambda)=C_p\lambda^{h-1}+O(\lambda^h) with an explicit alias coefficient CpC_p. For a fixed irreducible finite generator, holding times add a separate loss: the missing log-clock information is A2t2+O(t3)A_2t^2+O(t^3), or (A3+H)t3+O(t4)(A_3+H)t^3+O(t^4) when A2=0A_2=0. As a worked example we treat the stationary symmetric Ehrenfest chain of known size Ω\Omega observed at its endpoints, with sampling time tt and λ=Ωt\lambda=\Omega t. For every population and every finite positive tt, the exact endpoint information lies between an explicit quadratic lower bound and its matched Gaussian benchmark, Bn≤Rn<g(t)/λB_n\le R_n<g(t)/\lambda, with relative width at most 1/(2λ)1/(2\lambda) measured from BnB_n. The symmetric compound-Poisson benchmark is uniformly relatively accurate on 1≤λ≤LΩ1\le\lambda\le L_\Omega if and only if LΩ=o(Ω)L_\Omega=o(\Omega). A formula-level comparison with the binomial AR(1) literature (Cui–Lund; Weiß–Kim) has not been completed.

Provenance statement

OpenAI GPT-5.6 Sol, accessed through ChatGPT and Codex, was used to develop and audit the proofs and symbolic computations, and for drafting. Anthropic Claude Opus 5.5, used through Claude Code, was used to condense and edit the paper, verify the Gram-determinant algebra symbolically, and derive the random-walk alias corollary and the birth–death remark. The author reviewed and takes responsibility for every statement and proof.

Tools used

OpenAI
ChatGPTVersion GPT-5.6 Sol
OpenAI
CodexVersion GPT-5.6 Sol
Anthropic
Claude CodeVersion Claude Opus 5.5

References

  1. Duval, Céline; Hoffmann, Marc. Statistical inference across time scales. vol. 5, pp. 2004–2030. 2011DOI
  2. Duval, Céline. When is it no longer possible to estimate a compound Poisson process?. vol. 8, no. 1, pp. 274–301. 2014DOI
  3. Coca, Alberto J. Efficient nonparametric inference for discretely observed compound Poisson processes. vol. 170, pp. 475–523. 2018DOI
  4. Louis, Thomas A. Finding the observed information matrix when using the EM algorithm. vol. 44, no. 2, pp. 226–233. 1982DOI
  5. Bladt, Mogens; Sørensen, Michael. Statistical inference for discretely observed Markov jump processes. vol. 67, no. 3, pp. 395–410. 2005DOI
  6. Hobolth, Asger; Jensen, Jens Ledet. Statistical inference in evolutionary models of DNA sequences via the EM algorithm. vol. 4, no. 1. 2005DOI
  7. Hobolth, Asger; Jensen, Jens Ledet. Summary statistics for endpoint-conditioned continuous-time Markov chains. vol. 48, no. 4, pp. 911–924. 2011DOI
  8. Kurtz, Thomas G. Solutions of ordinary differential equations as limits of pure jump Markov processes. vol. 7, no. 1, pp. 49–58. 1970DOI
  9. Komorowski, Michał; Costa, Maria J.; Rand, David A.; Stumpf, Michael P. H. Sensitivity, robustness, and identifiability in stochastic chemical kinetics models. vol. 108, no. 21, pp. 8645–8650. 2011DOI
  10. Pantazis, Yannis; Katsoulakis, Markos A.; Vlachos, Dionisios G. Parametric sensitivity analysis for biochemical reaction networks based on pathwise information theory. vol. 14. 2013DOI
Show all 34 referencesShow fewer references
  1. Tataru, Paula; Hobolth, Asger. Comparison of methods for calculating conditional expectations of sufficient statistics for continuous time Markov chains. vol. 12. 2011DOI
  2. Radaelli, Marco; Landi, Gabriel T.; Modi, Kavan; Binder, Felix C. Fisher information of correlated stochastic processes. vol. 25. 2023DOI
  3. Fox, Zachary R.; Munsky, Brian. The finite state projection based Fisher information matrix approach to estimate information and optimize single-cell experiments. vol. 15, no. 1. 2019DOI
  4. Diaconis, Persi; Sturmfels, Bernd. Algebraic algorithms for sampling from conditional distributions. vol. 26, no. 1, pp. 363–397. 1998DOI
  5. Anderson, David F.; Craciun, Gheorghe; Kurtz, Thomas G. Product-form stationary distributions for deficiency zero chemical reaction networks. vol. 72, no. 8, pp. 1947–1970. 2010DOI
  6. Comte, Fabienne; Genon-Catalot, Valentine. Nonparametric adaptive estimation for pure jump Lévy processes. vol. 46, no. 3, pp. 595–617. 2010DOI
  7. Moor, Anne-Lena; Tjalma, Age; Reinhardt, Manuel; ten Wolde, Pieter Rein; Zechner, Christoph. State- versus Reaction-Based Information Processing in Biochemical Networks. vol. 136. 2026DOI
  8. Prech, Kacper; Landi, Gabriel T.; Meier, Florian; Nurgalieva, Nuriya; Potts, Patrick P.; Silva, Ralph; Mitchison, Mark T. Optimal Time Estimation and the Clock Uncertainty Relation for Stochastic Processes. vol. 15. 2025DOI
  9. Macieszczak, Katarzyna. Ultimate Kinetic Uncertainty Relation and Optimal Performance of Stochastic Clocks. 2024. PreprintDOI
  10. Di Terlizzi, Ivan; Baiesi, Marco. Kinetic uncertainty relation. vol. 52. 2019DOI
  11. Aït-Sahalia, Yacine; Jacod, Jean. Fisher's Information for Discretely Sampled Lévy Processes. vol. 76, no. 4, pp. 727–761. 2008DOI
  12. Diaconis, Persi; Griffiths, Robert. Exchangeable pairs of Bernoulli random variables, Krawtchouck polynomials, and Ehrenfest urns. vol. 54, no. 1, pp. 81–101. 2012DOI
  13. Weiß, Christian H.; Kim, Hee-Young. Parameter estimation for binomial AR(1) models with applications in finance and industry. vol. 54, pp. 563–590. 2013DOI
  14. Cui, Yunwei; Lund, Robert. Inference in binomial AR(1) models. vol. 80, no. 23–24, pp. 1985–1990. 2010DOI
  15. Pollard, David. A note on insufficiency and the preservation of Fisher information. From Probability to Statistics and Back: High-Dimensional Models and Processes–A Festschrift in Honor of Jon A. Wellner, vol. 9, pp. 266–275. 2013DOI
  16. Takeuchi, Kei; Takemura, Akimichi. On sum of 0–1 random variables. I. Univariate case. vol. 39, no. 1, pp. 85–102. 1987DOI
  17. Weiß, Christian H.; Kim, Hee-Young. Binomial AR(1) processes: moments, cumulants, and estimation. vol. 47, no. 3, pp. 494–510. 2013DOI
  18. Kingman, J. F. C. The imbedding problem for finite Markov chains. vol. 1, no. 1, pp. 14–24. 1962DOI
  19. Israel, Robert B.; Rosenthal, Jeffrey S.; Wei, Jason Z. Finding generators for Markov chains via empirical transition matrices, with applications to credit ratings. vol. 11, no. 2, pp. 245–265. 2001DOI
  20. Phillips, P. C. B. The problem of identification in finite parameter continuous time models. vol. 1, no. 4, pp. 351–362. 1973DOI
  21. Kalbfleisch, J. D.; Lawless, J. F. The analysis of panel data under a Markov assumption. vol. 80, no. 392, pp. 863–871. 1985DOI
  22. Becker, Gerhard; Kersting, Götz. Design problems for the pure birth process. vol. 15, no. 2, pp. 255–273. 1983DOI
  23. Pagendam, D. E.; Pollett, P. K. Locally optimal designs for the simple death process. vol. 140, no. 11, pp. 3096–3105. 2010DOI
  24. Aït-Sahalia, Yacine; Mykland, Per A. The effects of random and discrete sampling when estimating continuous-time diffusions. vol. 71, no. 2, pp. 483–549. 2003DOI

Version history

  1. v1 · 23 pagesSubmitted by Samuil PetkovInitial depositCurrentNov 19, 2026