# Talk:Volatility (finance)

WikiProject Finance (Rated C-class, Mid-importance)
This article is within the scope of WikiProject Finance, a collaborative effort to improve the coverage of articles related to Finance on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C  This article has been rated as C-Class on the project's quality scale.
Mid  This article has been rated as Mid-importance on the project's importance scale.
WikiProject Statistics (Rated C-class, Low-importance)

This article is within the scope of the WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page or join the discussion.

C  This article has been rated as C-Class on the quality scale.
Low  This article has been rated as Low-importance on the importance scale.

## Question

This does not sound right: future implied volatility which refers to the implied volatility observed from future prices of the financial instrument - that means I would need the future prices of the instrument to calculate the vola? :-) Where does all this terminology come from? Neither looks right nor backed by literature like the Hull. Thobitz (talk) 19:32, 22 March 2011 (UTC)

Is it really the standard deviation of the logarithmic returns, as the article says? Many online sources say it's the standard deviation of the (percentage) returns. Ben Finn (talk) 13:19, 18 December 2008 (UTC)

Yes - this is correct, at least from my context, which is options valuation. A most recent cite is "Option Pricing Models and Volatility Using Excel - VBA", by Rouah and Vainberg, Pg. 276. My firm generally uses this definition for finance research and consulting projects such as the valuation of employee stock options. The reason that the logarithmic definition is used is twofold:

1) The lower bound of a regular return is -100%. Using the logarithm of the return has no such limitation. This leads to...

2) Logarithmic returns are invertible. If a stock were to lose 5%, then gain 5%, (or vice-versa) it would not return to its initial price. But if the the returns are logarithmic, the math works out: for example: Given two returns r1 and r2, where log(r1)=5% and log(r2)=-5%, then the stock price after two days would be S2 = S * (1-r1) * (1-r2) = S * (e^0.05) * (e^-0.05) = S. This simplifies the models and calculations.

Catofgrey (talk) 21:46, 13 July 2009 (UTC)

The Lévy distribution, as far as I know, has infinite variance. Would "Lévy process" be more accurate? M C 999 (talk) 18:45, 16 June 2010 (UTC)

Mathematical definition?

Really? it sounds more as the definition and formulas related to annualized volatility. Not changing anything, waiting for discussion. Pablete85 (talk) 15:06, 24 May 2011 (UTC)

I'm a bit perplexed about the square root of time horizon used in the definition. Ok, I know units of measurement are often omitted in purely mathematical descriptions, but... It still seems to me like taking the square root of a time makes dimensionally no sense. Is the definition correct? Isn't there any trick which would make it dimensionally viable? Just asking, as a matter of fact I know nothing about the concepts involved. --Il wage (talk) 20:26, 1 April 2015 (UTC)

## standard deviation vs. semi deviation

The standard deviation is called volatility which is a measure of risk. semi deviation such as the downside risk can be defined as ${\displaystyle DR=E\left[\max \left\{E\left(R\right)-R,0\right\}\right]^{2}}$ while ${\displaystyle \sigma ^{2}=E\left[E\left(R\right)-R\right]^{2}}$. Jackzhp (talk) 02:42, 13 April 2010 (UTC)

## Crude Volatility Estimation section

"Of course, the average magnitude of the observations is merely an approximation of the standard deviation of the market index. Assuming that the market index daily changes are normally distributed with mean zero and standard deviation σ, the expected value of the magnitude of the observations is √(2/π)σ = 0.798σ. The net effect is that this crude approach overestimates the true volatility by about 25%."

I am no financier, but this last sentence does not seem correct. If the expectation of the observed deviations equals 0.798σ, then using this crude estimate rather than the true σ (= (1/.798)* average) would underestimate the true volatility, not overestimate it as stated.

--71.33.143.250 (talk) 21:46, 1 July 2011 (UTC)Doug P.S. Sorry if I inadvertently violated any standard - it's my first comment.

## I think the article is wrong

Several sources lately talk about Volatility Indices as a measure of "how much people are scared". Google it. --Athinker (talk) 00:43, 9 August 2011 (UTC)

## POV in criticism section

The criticism section seems awfully personal in the way it deals with critiquing volatility. Also, the papers cited are old. Updated references would be a good addition. JayBee51 (talk) 21:33, 30 August 2011 (UTC)

The criticism section does indeed seem a bit personal. That said, it would be unfair to ignore the critics. One point I disagree on: The age of the papers has no bearing on their relevance. In fact, old papers that are still cited are likely to be better than recent papers. I know this is different from the Wikipedia idea that newer is better; but, in academia, citations matter more than age. Cumulant (talk) 21:52, 31 May 2012 (UTC)

## Volatility is a Metric not a measure

I think the opening section is not correct, and leads to confusion further into the article. "Volatility" is a concept for randomness - but upside and downside. This fits the definition of a "Risk Metric". It can be calculated in a number of different ways - (option implied, exponentially weighted moving average etc). All these are risk measures.

Making this distinction early on in the piece would make the definitions of the various risk measures (calculation algorithms) more meaningful. — Preceding unsigned comment added by 168.140.181.4 (talk) 02:34, 20 August 2013 (UTC)

Dr. Wu has reviewed this Wikipedia page, and provided us with the following comments to improve its quality:

The section on "Estimate of compound annual growth rate (CAGR)" and "Implied Volatility parametrisation" do not seem to fit. "Crude volatility estimation" is also a strange section.

Instead, there should be a chapter introducing volatility forecasting models such as GARCH models, the RiskMetrics model, before talking about "Criticisms of volatility forecasting models"

The criticism from Taleb is a bit cynical and not constructive.

We hope Wikipedians on this talk page can take advantage of these comments and improve the quality of the article accordingly.

Dr. Wu has published scholarly research which seems to be relevant to this Wikipedia article:

• Reference : Jing-zhi Huang & Liuren Wu, 2004. "Specification Analysis of Option Pricing Models Based on Time-Changed Levy Processes," Econometric Society 2004 North American Winter Meetings 405, Econometric Society.

ExpertIdeasBot (talk) 04:55, 16 June 2016 (UTC)

Dr. Kocenda has reviewed this Wikipedia page, and provided us with the following comments to improve its quality:

this comment relates to the realized volatility/the realized variance and its connection to the direction of volatility:

Realized variance can be decomposed into directions via realized semivariances. Realized semivariances were introduced by Barndorff-Nielsen et al. (2010) - they enable to isolate and capture negative and positive shocks to volatility, hence its direction. They are also ideally suited to interpret qualitative differences in volatility spillovers. Extent of volatility spillovers is manifested via connectedness that is measured with the Diebold-Yilmaz spillover index - DY index (Diebold and Yilmaz, 2015). DY index uses realized variance and measures total, directional and net connectedness. Barunik et al. (2016) use realized semivariances to measure asymmetric connectedness that quantifies asymmetries in spillovers, their directions etc.

References: Barndorff-Nielsen, O., S. Kinnebrock, and N. Shephard (2010). Volatility and Time Series Econometrics: Essays in Honor of Robert F. Engle, Chapter Measuring Downside Risk-Realised Semivariance. Oxford University Press. Diebold, F. X. and K. Yilmaz (2015). Financial and Macroeconomic Connectedness: A Network Approach to Measurement and Monitoring. Oxford University Press, USA.Barunik, J., E. Kocenda, and L. Vacha (2016). Asymmetric connectedness on the US stock market: Bad and good volatility spillovers. Journal of Financial Markets 27, 55-78.

We hope Wikipedians on this talk page can take advantage of these comments and improve the quality of the article accordingly.

Dr. Kocenda has published scholarly research which seems to be relevant to this Wikipedia article:

• Reference : Evzen Kocenda & Vit Bubak & Filip Zikes, 2011. "Volatility Transmission in Emerging European Foreign Exchange Markets," William Davidson Institute Working Papers Series wp1020, William Davidson Institute at the University of Michigan.

ExpertIdeasBot (talk) 18:28, 27 June 2016 (UTC)

Dr. Caporin has reviewed this Wikipedia page, and provided us with the following comments to improve its quality:

I would change the sentence

"It is the volatility of a financial instrument based on historical prices over the specified period with the last observation the most recent price." Into "The actual current volatility of a financial instrument is generally estimated starting from price data recovered over the specified period. The volatility is measured from price returns."

I would put something before "Actual historical volatility" something like

"...of an instrument, and other definitions of volatility:"

The definition of realized volatility is inappropriate as realized volatility is generally estimated by means of data at a hgher frequency compared to that adopted for traditional volatility measures.

The definitions of future volatility measures is inappropriate as future prices are not observable. These measures are recovered from forecasts of future prices.

I do not agree with definition 1 for volatility for investors as, from a behavioral perspective, I do think that investors worry more if volatility is higher

I would add in 6 "Price volatility presents risk of buying assets overprices and sell them when underpriced"

I do not agree with "Most typically, extreme movements do not appear 'out of nowhere'; they are presaged by larger movements than usual. This is termed autoregressive conditional heteroskedasticity."

The reference to "autoregressive conditional heteroskedasticity" would be better placed above

"It is common knowledge that types of assets experience periods of high and low volatility. That is, during some periods, prices go up and down quickly, while during other times they barely move at all."

"This is commonly referred to as volatility clustering and lead to the development of statistical models starting from the autoregressive conditional heteroskedasticity approach of Engle"

Add a reference to the seminal work of the nobel price Engle, published in 1982

Moreover, burst of volatility can appear "out of nowhere" when we have an unexpected market shock

In Estimate of compound annual growth rate (CAGR) you have not defined AR nor sigma as in the first equation they are not present

We hope Wikipedians on this talk page can take advantage of these comments and improve the quality of the article accordingly.

We believe Dr. Caporin has expertise on the topic of this article, since he has published relevant scholarly research:

• Reference : Massimiliano Caporin & Eduardo Rossi & Paolo Santucci de Magistris, 2014. "Chasing volatility - A persistent multiplicative error model with jumps," CREATES Research Papers 2014-29, School of Economics and Management, University of Aarhus.

ExpertIdeasBot (talk) 16:08, 24 August 2016 (UTC)