pith. sign in

arxiv: 2605.13607 · v1 · pith:V5Q3Y4IMnew · submitted 2026-05-13 · 📊 stat.CO · cs.CE· cs.MS

Ergodicity Library: A Python Toolkit for Stochastic-Process Simulation, Time-Average Diagnostics, and Agent-Based Experiments

Ihor Kendiukhov This is my paper

Pith reviewed 2026-06-30 21:13 UTC · model grok-4.3

classification 📊 stat.CO cs.CEcs.MS
keywords ergodicitystochastic processesPython librarynon-ergodicitytime averagesagent-based modelsheavy-tailed processessimulation toolkit
0
0 comments X

The pith

An open-source Python library unifies stochastic-process simulators, time-average diagnostics, and agent-based experiments.

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

The paper presents the Ergodicity Library as a toolkit that combines process definitions and simulators, analysis and fitting tools, and agent-based experimentation in one package. These elements are typically scattered across separate scripts, so the library aims to reduce the glue code needed to go from specifying a process to running diagnostics and experiments. A sympathetic reader would care because it supports work on non-ergodicity, heavy-tailed processes, and decision making under uncertainty with reproducible examples. The library is built on the scientific Python stack and documents its architecture, supported processes, and practical limits rather than new theory.

Core claim

The Ergodicity Library is positioned as an integration layer on top of the scientific Python stack that brings together process definitions and simulators, analysis and fitting tools, and agent-based experimentation for computational work on stochastic dynamics with emphasis on non-ergodicity, time-average behavior, heavy-tailed processes, and decision making under uncertainty.

What carries the argument

The three-layer architecture of process definitions and simulators, analysis and fitting tools, and agent-based experimentation.

If this is right

  • Users can perform heavy-tailed ensemble spread analysis and multiplicative Levy growth diagnostics from a single workflow.
  • The package supports adaptive memory mean reversion studies and preasymptotic fluctuation analysis without separate scripts.
  • Partial stochastic differential equation simulations become available alongside agent-based comparative experiments.
  • The amount of custom glue code required to move from process specification to diagnostics is reduced.

Where Pith is reading between the lines

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

  • The unified interface could enable direct side-by-side testing of different stochastic models in decision-making scenarios that previously required extensive custom coding.
  • Researchers studying time-average properties might now more easily incorporate agent-based elements into their workflows.
  • The library's structure suggests it could serve as a base for community contributions of additional process families or diagnostic methods.

Load-bearing premise

The implemented simulators and diagnostics correctly reproduce the intended stochastic behaviors and time-average properties without numerical artifacts or implementation errors that would affect the documented examples.

What would settle it

Running the fully reproducible examples in the package and obtaining ensemble spreads, time averages, or fluctuation statistics that deviate systematically from known theoretical expectations for the tested heavy-tailed or multiplicative processes.

Figures

Figures reproduced from arXiv: 2605.13607 by Ihor Kendiukhov.

Figure 1
Figure 1. Figure 1: Quantile fan charts for Brownian and Levy-stable ensembles. The heavy-tailed [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Trajectory intermittency and summary-statistic divergence for a Geometric Levy [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Adaptive-rate Ornstein–Uhlenbeck dynamics compared with a fixed-rate baseline. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Preasymptotic convergence and fluctuation diagnostics for log-wealth under Geo [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Partial stochastic differential equation simulation with a contour view and spatial [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
read the original abstract

ergodicity is an open-source Python library for computational work on stochastic dynamics, with particular emphasis on non-ergodicity, time-average behavior, heavy-tailed processes, and decision making under uncertainty. The package brings together three layers that are often split across ad hoc scripts: process definitions and simulators, analysis and fitting tools, and agent-based experimentation. This article documents the implemented software rather than presenting new stochastic theory. We describe the package architecture, the supported process families, the analysis workflow, and the practical boundaries of the current implementation. We also provide fully reproducible examples covering heavy-tailed ensemble spread, multiplicative Levy growth diagnostics, adaptive memory mean reversion, preasymptotic fluctuation analysis, and partial stochastic differential equation simulation. The package is positioned as an integration layer on top of the scientific Python stack, reducing the amount of glue code required to move from process specification to diagnostics and comparative experiments.

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 / 2 minor

Summary. The manuscript describes the open-source Python library 'ergodicity' for computational work on stochastic dynamics, with emphasis on non-ergodicity, time-average behavior, heavy-tailed processes, and decision making under uncertainty. It documents the package architecture, supported process families, analysis workflow, and practical boundaries of the implementation rather than presenting new theory. The work highlights the integration of process simulators, analysis/fitting tools, and agent-based experimentation, and provides fully reproducible examples covering heavy-tailed ensemble spread, multiplicative Levy growth diagnostics, adaptive memory mean reversion, preasymptotic fluctuation analysis, and partial stochastic differential equation simulation. The package is positioned as an integration layer atop the scientific Python stack to reduce glue code.

Significance. If the implementations are correct as documented, the library would offer a practical consolidation of tools that are frequently scattered across ad hoc scripts, potentially streamlining workflows for researchers studying time averages and non-ergodic stochastic processes. The explicit provision of reproducible examples is a clear strength that supports usability and verifiability of the described features.

major comments (2)
  1. Abstract: The central claim that the package integrates three layers (simulators, analysis tools, and agent-based experiments) that are often split across ad hoc scripts is presented descriptively but without any concrete illustration or metric showing reduced glue code or workflow simplification, leaving the integration benefit unquantified.
  2. Section describing the practical boundaries of the current implementation: The boundaries are referenced but lack specific details on numerical stability, performance limits, or known discrepancies between simulated time averages and theoretical expectations, which directly affects assessment of the reliability of the documented diagnostics.
minor comments (2)
  1. The manuscript would benefit from a summary table listing all supported process families with their key parameters and diagnostic capabilities for quick reference.
  2. Ensure that each reproducible example section explicitly cross-references the corresponding code location or script name to facilitate immediate verification by readers.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments and the recommendation for minor revision. We respond to each major comment below.

read point-by-point responses

  1. Referee: Abstract: The central claim that the package integrates three layers (simulators, analysis tools, and agent-based experiments) that are often split across ad hoc scripts is presented descriptively but without any concrete illustration or metric showing reduced glue code or workflow simplification, leaving the integration benefit unquantified.

    Authors: The manuscript states that the library reduces glue code by integrating the three layers and supports this with fully reproducible examples that show complete workflows. We acknowledge, however, that the abstract and main text do not include explicit quantitative metrics (such as code-length comparisons or timing benchmarks against ad-hoc scripts). We will revise the abstract and add a short illustrative comparison in the examples section to quantify the integration benefit where feasible. revision: yes

  2. Referee: Section describing the practical boundaries of the current implementation: The boundaries are referenced but lack specific details on numerical stability, performance limits, or known discrepancies between simulated time averages and theoretical expectations, which directly affects assessment of the reliability of the documented diagnostics.

    Authors: The manuscript references the practical boundaries of the implementation but does not supply the requested quantitative details on numerical stability, performance limits, or discrepancies with theory. We agree that these specifics would strengthen the reliability assessment. We will expand the relevant section with concrete information drawn from the library's implementation and validation tests. revision: yes

Circularity Check

0 steps flagged

No circularity: software documentation without derivations

full rationale

The manuscript is a software description paper that documents an open-source Python library for stochastic-process simulation and diagnostics. It explicitly states it presents no new stochastic theory or derivations. No equations, fitted parameters, predictions, or load-bearing self-citations appear that could reduce to inputs by construction. The central claim is the integration of existing components into a toolkit, which is independent of any circular reduction. This is the standard case of a self-contained library paper with no mathematical claims to inspect for circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The paper introduces no free parameters, axioms, or invented entities because it presents no new theoretical derivations or mathematical claims.

pith-pipeline@v0.9.1-grok · 5689 in / 940 out tokens · 25591 ms · 2026-06-30T21:13:32.424279+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

2 extracted references · 2 canonical work pages

  1. [1]

    Re-Run, Repeat, Reproduce, Reuse, Replicate: Trans- forming Code into Scientific Contributions

    Benureau FCY, Rougier NP (2018). “Re-Run, Repeat, Reproduce, Reuse, Replicate: Trans- forming Code into Scientific Contributions.”Frontiers in Neuroinformatics,11, 69.doi: 10.3389/fninf.2017.00069. Bonabeau E (2002). “Agent-Based Modeling: Methods and Techniques for Simulating Human Systems.”Proceedings of the National Academy of Sciences,99(suppl_3), 728...

    work page doi:10.3389/fninf.2017.00069 2018

  2. [2]

    A Method for Simulating Stable Random Variables

    Mechanisms, Models and Physical Applications.”Physics Reports,195(4–5), 127–293. doi:10.1016/0370-1573(90)90099-N. Ihor Kendiukhov9 Chambers JM, Mallows CL, Stuck BW (1976). “A Method for Simulating Stable Random Variables.”Journal of the American Statistical Association,71(354), 340–344.doi:10. 1080/01621459.1976.10480344. Cont R (2001). “Empirical Prope...

    work page doi:10.1016/0370-1573(90)90099-n 1976