dorsal/arxiv
View SchemaNonstationary Increments, Scaling Distributions, and Variable Diffusion Processes in Financial Markets
| Authors | Kevin E. Bassler, Joseph L. McCauley, Gemunu H. Gunaratne |
|---|---|
| Categories | |
| ArXiv ID | physics/0609198 |
| URL | https://arxiv.org/abs/physics/0609198 |
| DOI | 10.1073/pnas.0708664104 |
Abstract
Arguably the most important problem in quantitative finance is to understand the nature of stochastic processes that underlie market dynamics. One aspect of the solution to this problem involves determining characteristics of the distribution of fluctuations in returns. Empirical studies conducted over the last decade have reported that they arenon-Gaussian, scale in time, and have power-law(or fat) tails. However, because they use sliding interval methods of analysis, these studies implicitly assume that the underlying process has stationary increments. We explicitly show that this assumption is not valid for the Euro-Dollar exchange rate between 1999-2004. In addition, we find that fluctuations in returns of the exchange rate are uncorrelated and scale as power-laws for certain time intervals during each day. This behavior is consistent with a diffusive process with a diffusion coefficient that depends both on the time and the price change. Within scaling regions, we find that sliding interval methods can generate fat-tailed distributions as an artifact, and that the type of scaling reported in many previous studies does not exist.
{
"annotation_id": "12a73b4b-0f1a-49bd-acff-ea3bec0f6609",
"date_created": "2026-03-02T18:01:14.465000Z",
"date_modified": "2026-03-02T18:01:14.465000Z",
"file_hash": "0f2415a56e726a32102ad2dd5083b17488be73e4b591e32bf224c415d10680c8",
"private": false,
"record": {
"abstract": "Arguably the most important problem in quantitative finance is to understand\nthe nature of stochastic processes that underlie market dynamics. One aspect of\nthe solution to this problem involves determining characteristics of the\ndistribution of fluctuations in returns. Empirical studies conducted over the\nlast decade have reported that they arenon-Gaussian, scale in time, and have\npower-law(or fat) tails. However, because they use sliding interval methods of\nanalysis, these studies implicitly assume that the underlying process has\nstationary increments. We explicitly show that this assumption is not valid for\nthe Euro-Dollar exchange rate between 1999-2004. In addition, we find that\nfluctuations in returns of the exchange rate are uncorrelated and scale as\npower-laws for certain time intervals during each day. This behavior is\nconsistent with a diffusive process with a diffusion coefficient that depends\nboth on the time and the price change. Within scaling regions, we find that\nsliding interval methods can generate fat-tailed distributions as an artifact,\nand that the type of scaling reported in many previous studies does not exist.",
"arxiv_id": "physics/0609198",
"authors": [
"Kevin E. Bassler",
"Joseph L. McCauley",
"Gemunu H. Gunaratne"
],
"categories": [
"physics.soc-ph",
"cond-mat.stat-mech",
"physics.data-an",
"q-fin.ST"
],
"doi": "10.1073/pnas.0708664104",
"title": "Nonstationary Increments, Scaling Distributions, and Variable Diffusion Processes in Financial Markets",
"url": "https://arxiv.org/abs/physics/0609198"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "b73f9f6c-5c22-426e-8da7-50163925cc32",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}