dorsal/arxiv
View SchemaRoughness and Finite Size Effect in the NYSE Stock-Price Fluctuations
| Authors | V. Alfi, F. Coccetti, A. Petri, L. Pietronero |
|---|---|
| Categories | |
| ArXiv ID | physics/0602052 |
| URL | https://arxiv.org/abs/physics/0602052 |
| DOI | 10.1140/epjb/e2006-00240-8 |
Abstract
We consider the roughness properties of NYSE (New York Stock Exchange) stock-price fluctuations. The statistical properties of the data are relatively homogeneous within the same day but the large jumps between different days prevent the extension of the analysis to large times. This leads to intrinsic finite size effects which alter the apparent Hurst (H) exponent. We show, by analytical methods, that finite size effects always lead to an enhancement of H. We then consider the effect of fat tails on the analysis of the roughness and show that the finite size effects are strongly enhanced by the fat tails. The non stationarity of the stock price dynamics also enhances the finite size effects which, in principle, can become important even in the asymptotic regime. We then compute the Hurst exponent for a set of stocks of the NYSE and argue that the interpretation of the value of H is highly ambiguous in view of the above results. Finally we propose an alternative determination of the roughness in terms of the fluctuations from moving averages with variable characteristic times. This permits to eliminate most of the previous problems and to characterize the roughness in useful way. In particular this approach corresponds to the automatic elimination of trends at any scale.
{
"annotation_id": "aa75a0f8-e9b3-4d20-89dc-8a8149b3d8b7",
"date_created": "2026-03-02T18:01:06.551000Z",
"date_modified": "2026-03-02T18:01:06.551000Z",
"file_hash": "15f7eb373e8421c569791bfdc76f8215e20cd15c2e34b644df7bfeff7cd23f7c",
"private": false,
"record": {
"abstract": "We consider the roughness properties of NYSE (New York Stock Exchange)\nstock-price fluctuations. The statistical properties of the data are relatively\nhomogeneous within the same day but the large jumps between different days\nprevent the extension of the analysis to large times. This leads to intrinsic\nfinite size effects which alter the apparent Hurst (H) exponent. We show, by\nanalytical methods, that finite size effects always lead to an enhancement of\nH. We then consider the effect of fat tails on the analysis of the roughness\nand show that the finite size effects are strongly enhanced by the fat tails.\nThe non stationarity of the stock price dynamics also enhances the finite size\neffects which, in principle, can become important even in the asymptotic\nregime. We then compute the Hurst exponent for a set of stocks of the NYSE and\nargue that the interpretation of the value of H is highly ambiguous in view of\nthe above results. Finally we propose an alternative determination of the\nroughness in terms of the fluctuations from moving averages with variable\ncharacteristic times. This permits to eliminate most of the previous problems\nand to characterize the roughness in useful way. In particular this approach\ncorresponds to the automatic elimination of trends at any scale.",
"arxiv_id": "physics/0602052",
"authors": [
"V. Alfi",
"F. Coccetti",
"A. Petri",
"L. Pietronero"
],
"categories": [
"physics.soc-ph",
"q-fin.ST"
],
"doi": "10.1140/epjb/e2006-00240-8",
"title": "Roughness and Finite Size Effect in the NYSE Stock-Price Fluctuations",
"url": "https://arxiv.org/abs/physics/0602052"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "077c75aa-3912-460c-b294-780c2bd0fc88",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}