dorsal/arxiv
View SchemaNoise sensitivity of portfolio selection under various risk measures
| Authors | Imre Kondor, Szilard Pafka, Gabor Nagy |
|---|---|
| Categories | |
| ArXiv ID | physics/0611027 |
| URL | https://arxiv.org/abs/physics/0611027 |
Abstract
We study the sensitivity to estimation error of portfolios optimized under various risk measures, including variance, absolute deviation, expected shortfall and maximal loss. We introduce a measure of portfolio sensitivity and test the various risk measures by considering simulated portfolios of varying sizes N and for different lengths T of the time series. We find that the effect of noise is very strong in all the investigated cases, asymptotically it only depends on the ratio N/T, and diverges at a critical value of N/T, that depends on the risk measure in question. This divergence is the manifestation of a phase transition, analogous to the algorithmic phase transitions recently discovered in a number of hard computational problems. The transition is accompanied by a number of critical phenomena, including the divergent sample to sample fluctuations of portfolio weights. While the optimization under variance and mean absolute deviation is always feasible below the critical value of N/T, expected shortfall and maximal loss display a probabilistic feasibility problem, in that they can become unbounded from below already for small values of the ratio N/T, and then no solution exists to the optimization problem under these risk measures. Although powerful filtering techniques exist for the mitigation of the above instability in the case of variance, our findings point to the necessity of developing similar filtering procedures adapted to the other risk measures where they are much less developed or nonexistent. Another important message of this study is that the requirement of robustness (noise-tolerance) should be given special attention when considering the theoretical and practical criteria to be imposed on a risk measure.
{
"annotation_id": "03e1287c-d6f1-4cb5-a15c-de5b7026193a",
"date_created": "2026-03-02T18:01:14.147000Z",
"date_modified": "2026-03-02T18:01:14.147000Z",
"file_hash": "85d72c551eabbbe60b5e3d2e078b08a348e6e4c1d8a82475bc308dfa24d0fdc7",
"private": false,
"record": {
"abstract": "We study the sensitivity to estimation error of portfolios optimized under\nvarious risk measures, including variance, absolute deviation, expected\nshortfall and maximal loss. We introduce a measure of portfolio sensitivity and\ntest the various risk measures by considering simulated portfolios of varying\nsizes N and for different lengths T of the time series. We find that the effect\nof noise is very strong in all the investigated cases, asymptotically it only\ndepends on the ratio N/T, and diverges at a critical value of N/T, that depends\non the risk measure in question. This divergence is the manifestation of a\nphase transition, analogous to the algorithmic phase transitions recently\ndiscovered in a number of hard computational problems. The transition is\naccompanied by a number of critical phenomena, including the divergent sample\nto sample fluctuations of portfolio weights. While the optimization under\nvariance and mean absolute deviation is always feasible below the critical\nvalue of N/T, expected shortfall and maximal loss display a probabilistic\nfeasibility problem, in that they can become unbounded from below already for\nsmall values of the ratio N/T, and then no solution exists to the optimization\nproblem under these risk measures. Although powerful filtering techniques exist\nfor the mitigation of the above instability in the case of variance, our\nfindings point to the necessity of developing similar filtering procedures\nadapted to the other risk measures where they are much less developed or\nnonexistent. Another important message of this study is that the requirement of\nrobustness (noise-tolerance) should be given special attention when considering\nthe theoretical and practical criteria to be imposed on a risk measure.",
"arxiv_id": "physics/0611027",
"authors": [
"Imre Kondor",
"Szilard Pafka",
"Gabor Nagy"
],
"categories": [
"physics.soc-ph",
"q-fin.RM"
],
"title": "Noise sensitivity of portfolio selection under various risk measures",
"url": "https://arxiv.org/abs/physics/0611027"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "c2263eb4-426b-49f2-ace8-d03d6b0f6613",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}