Volatility is log-normal -- but not for the reason you think

Martin Tegnér; Rolf Poulsen

doi:10.3390/risks6020046

Volatility is log-normal -- but not for the reason you think

Martin Tegnér, Rolf Poulsen

Department of Mathematical Sciences

Research output: Contribution to journal › Journal article › Research › peer-review

8 Citations (Scopus)

289 Downloads (Pure)

Abstract

It is impossible to discriminate between the commonly used stochastic volatility models of Heston, log-normal, and 3-over-2 on the basis of exponentially weighted averages of daily returns—even though it appears so at first sight. However, with a 5-min sampling frequency, the models can be differentiated and empirical evidence overwhelmingly favours a fast mean-reverting log-normal model.

Original language	English
Article number	46
Journal	Risks
Volume	6
Issue number	2
Number of pages	16
ISSN	2227-9091
DOIs	https://doi.org/10.3390/risks6020046
Publication status	Published - Jun 2018

Access to Document

10.3390/risks6020046Licence: CC BY

Volatility Is Log-Normal—But Not for the Reason You ThinkFinal published version, 1.09 MBLicence: CC BY

Cite this

@article{2ca0d9757cfc4d48a58a409dba1d5ab5,

title = "Volatility is log-normal -- but not for the reason you think",

abstract = "It is impossible to discriminate between the commonly used stochastic volatility models of Heston, log-normal, and 3-over-2 on the basis of exponentially weighted averages of daily returns—even though it appears so at first sight. However, with a 5-min sampling frequency, the models can be differentiated and empirical evidence overwhelmingly favours a fast mean-reverting log-normal model. ",

author = "Martin Tegn{\'e}r and Rolf Poulsen",

year = "2018",

month = jun,

doi = "10.3390/risks6020046",

language = "English",

volume = "6",

journal = "Risks",

issn = "2227-9091",

publisher = "MDPI",

number = "2",

}

TY - JOUR

T1 - Volatility is log-normal -- but not for the reason you think

AU - Tegnér, Martin

AU - Poulsen, Rolf

PY - 2018/6

Y1 - 2018/6

N2 - It is impossible to discriminate between the commonly used stochastic volatility models of Heston, log-normal, and 3-over-2 on the basis of exponentially weighted averages of daily returns—even though it appears so at first sight. However, with a 5-min sampling frequency, the models can be differentiated and empirical evidence overwhelmingly favours a fast mean-reverting log-normal model.

AB - It is impossible to discriminate between the commonly used stochastic volatility models of Heston, log-normal, and 3-over-2 on the basis of exponentially weighted averages of daily returns—even though it appears so at first sight. However, with a 5-min sampling frequency, the models can be differentiated and empirical evidence overwhelmingly favours a fast mean-reverting log-normal model.

U2 - 10.3390/risks6020046

DO - 10.3390/risks6020046

M3 - Journal article

VL - 6

JO - Risks

JF - Risks

SN - 2227-9091

IS - 2

M1 - 46

ER -