由买买提看人间百态

boards

本页内容为未名空间相应帖子的节选和存档,一周内的贴子最多显示50字,超过一周显示500字 访问原贴
Quant版 - 问个stochastic volatility model calibration的问题
相关主题
请问heston model怎么hedge?MFE deadline的问题
Yahoo Finance数据源一问弱问一题,关于binomial的
[合集] 面试问题请教成绩单问题-请问CMU在读同学,或今年的申请人
战五渣再问个implied volatility的问题Pricing a trinomial call option
问个volatility surface的问题还是分红和期权的问题
Heston model calibration再借人气问一个关于期权Cross Sectional Data的问题
暑期实习求推荐。。关于选Offer。。。大家来给看看
求教Bessel过程问个绿书上的问题 - delta hedging
相关话题的讨论汇总
话题: sv话题: model话题: market话题: hedging话题: sdf
进入Quant版参与讨论
1 (共1页)
q*****l
发帖数: 124
1
请问一般来说SV model都是用option的market price来做calibration的,然而同样的我
们貌似也可以用market price of stock来直接估计SV model的参数。
请问这两种方法哪个更普遍一点?
p******i
发帖数: 1358
2
use option price
what you need is the RN parameter.

【在 q*****l 的大作中提到】
: 请问一般来说SV model都是用option的market price来做calibration的,然而同样的我
: 们貌似也可以用market price of stock来直接估计SV model的参数。
: 请问这两种方法哪个更普遍一点?

d**0
发帖数: 124
3
option price, implied vol和underlying price没关系

【在 q*****l 的大作中提到】
: 请问一般来说SV model都是用option的market price来做calibration的,然而同样的我
: 们貌似也可以用market price of stock来直接估计SV model的参数。
: 请问这两种方法哪个更普遍一点?

C******n
发帖数: 9204
4
果然是没区分Physical和RN的。。。直接用model-free的方法吧,VIX那种的。
L*******t
发帖数: 2385
5
要看你的模型了,很多SV加Jump的模型都没办法直接用option price calibrate
同样的问题我在版上问过,但是没人回答。。
另一个问题是RN measure和physical measure的参数到底有什么联系,这个完全看你的
SPD
由于SVmodel是incomplete的,你完全可以选到一个RN measure使得physical的参数和
RN参数大多数都是一样的。

【在 q*****l 的大作中提到】
: 请问一般来说SV model都是用option的market price来做calibration的,然而同样的我
: 们貌似也可以用market price of stock来直接估计SV model的参数。
: 请问这两种方法哪个更普遍一点?

C******n
发帖数: 9204
6
两者必须是不一样的。或者这么说吧,RN和phy差别在哪里?(不要告诉我等价鞅测度)

【在 L*******t 的大作中提到】
: 要看你的模型了,很多SV加Jump的模型都没办法直接用option price calibrate
: 同样的问题我在版上问过,但是没人回答。。
: 另一个问题是RN measure和physical measure的参数到底有什么联系,这个完全看你的
: SPD
: 由于SVmodel是incomplete的,你完全可以选到一个RN measure使得physical的参数和
: RN参数大多数都是一样的。

j*****4
发帖数: 292
7
change measure跟vol有关系吗?

【在 C******n 的大作中提到】
: 果然是没区分Physical和RN的。。。直接用model-free的方法吧,VIX那种的。
C******n
发帖数: 9204
8
我觉得Heston已经解释的非常清楚了吧。ultimately, 为什么会有RN prob/measure这
么个东西?economic meaning是什么?(不必要说只是数学上方便,更不要说只有在RN
下才可以解)

【在 j*****4 的大作中提到】
: change measure跟vol有关系吗?
p******i
发帖数: 1358
9
I don't understand
for SV+Jump model, the RN measure and the physical one have no link at all
(not like GBM, in which the sigma is the same for both measures)
Of course you can pick some parameters as you like and try to fit the rest,
but what is the point? I bet your model will fit the market very poorly.

【在 L*******t 的大作中提到】
: 要看你的模型了,很多SV加Jump的模型都没办法直接用option price calibrate
: 同样的问题我在版上问过,但是没人回答。。
: 另一个问题是RN measure和physical measure的参数到底有什么联系,这个完全看你的
: SPD
: 由于SVmodel是incomplete的,你完全可以选到一个RN measure使得physical的参数和
: RN参数大多数都是一样的。

C******n
发帖数: 9204
10
你不能说因为 "this model"是Incomplete,“所以”你就可以任意选一个SDF。
对于你所感兴趣的market,例如option market,fact会告诉你SDF应该长什么样。
financial market大体上是complete的,虽然有很多的noise,所谓incomplete只是因
为你的model focus on 1 asset and 2 or 3 shocks,这只是你的modeling technique
的限制,不代表你应该这么反映现实。
举个例子,BS + SV或者BS + jump都是incomplete,但merton和heston都是怎么确定SDF
的,也就是说他们做了何种假设?是否合理?不合理的话现在大家在干什么?

【在 L*******t 的大作中提到】
: 要看你的模型了,很多SV加Jump的模型都没办法直接用option price calibrate
: 同样的问题我在版上问过,但是没人回答。。
: 另一个问题是RN measure和physical measure的参数到底有什么联系,这个完全看你的
: SPD
: 由于SVmodel是incomplete的,你完全可以选到一个RN measure使得physical的参数和
: RN参数大多数都是一样的。

相关主题
Heston model calibrationMFE deadline的问题
暑期实习求推荐。。弱问一题,关于binomial的
求教Bessel过程成绩单问题-请问CMU在读同学,或今年的申请人
进入Quant版参与讨论
s******e
发帖数: 1751
11
BS + SV or BS + Jump market is incomplete.
if market is incomplete, i.e. nobody is trading on those parameters, why do
you bother w/ those models?

technique
SDF

【在 C******n 的大作中提到】
: 你不能说因为 "this model"是Incomplete,“所以”你就可以任意选一个SDF。
: 对于你所感兴趣的market,例如option market,fact会告诉你SDF应该长什么样。
: financial market大体上是complete的,虽然有很多的noise,所谓incomplete只是因
: 为你的model focus on 1 asset and 2 or 3 shocks,这只是你的modeling technique
: 的限制,不代表你应该这么反映现实。
: 举个例子,BS + SV或者BS + jump都是incomplete,但merton和heston都是怎么确定SDF
: 的,也就是说他们做了何种假设?是否合理?不合理的话现在大家在干什么?

C******n
发帖数: 9204
12
The actual market并没有模型里所说的incompleteness。incomplete的是stupid
model。换句话说,这是个模型的缺陷。我们需要做假设来弥补。
actual market上有Jump和SV吗?市场还在吗?

do

【在 s******e 的大作中提到】
: BS + SV or BS + Jump market is incomplete.
: if market is incomplete, i.e. nobody is trading on those parameters, why do
: you bother w/ those models?
:
: technique
: SDF

s******e
发帖数: 1751
13
market could be incomplete. (i want to trade bitcoin calls. who is going to
make a market for me?)
given that, what's the point of talking about those fancy model (there is a
point for some products, but that's very nichy, to which most of noices do
not have exposure)?
there are jump and sv on the market (credit risk and VIX vol). both are big
market.

【在 C******n 的大作中提到】
: The actual market并没有模型里所说的incompleteness。incomplete的是stupid
: model。换句话说,这是个模型的缺陷。我们需要做假设来弥补。
: actual market上有Jump和SV吗?市场还在吗?
:
: do

x********o
发帖数: 519
14
do you forgot about risk management? in many cases you have to have SV
models, or even SV + jumps

to
a
big

【在 s******e 的大作中提到】
: market could be incomplete. (i want to trade bitcoin calls. who is going to
: make a market for me?)
: given that, what's the point of talking about those fancy model (there is a
: point for some products, but that's very nichy, to which most of noices do
: not have exposure)?
: there are jump and sv on the market (credit risk and VIX vol). both are big
: market.

x********o
发帖数: 519
15
in many cases you cannot pick up the parameters arbitrarily.
just take the Heston model for example, if you ever use this model for MC
simulation, you will know the real issues on the range of the parameters you
can choose.

,

【在 p******i 的大作中提到】
: I don't understand
: for SV+Jump model, the RN measure and the physical one have no link at all
: (not like GBM, in which the sigma is the same for both measures)
: Of course you can pick some parameters as you like and try to fit the rest,
: but what is the point? I bet your model will fit the market very poorly.

C******n
发帖数: 9204
16
market肯定是incomplete的,没人能买卖任意数量。
在SV model里是incomplete,但这是this model的缺陷,因为实际上有比该模型更多的
product存在但没考虑。Potentially,任意的shock都可以write contract。正如你所
说,vol是个market,而且liquidity足够好,for the purpose of SV modeling,可以
算是complete。
我主要说的是上面童鞋想模型有点太literal,model说incomplete就能任意设置SDF,
这是不合理的。

to
a
big

【在 s******e 的大作中提到】
: market could be incomplete. (i want to trade bitcoin calls. who is going to
: make a market for me?)
: given that, what's the point of talking about those fancy model (there is a
: point for some products, but that's very nichy, to which most of noices do
: not have exposure)?
: there are jump and sv on the market (credit risk and VIX vol). both are big
: market.

p******i
发帖数: 1358
17
exactly,
incomplete market means THE MARKET will choose one of the equivalent RN
measure not you

you

【在 x********o 的大作中提到】
: in many cases you cannot pick up the parameters arbitrarily.
: just take the Heston model for example, if you ever use this model for MC
: simulation, you will know the real issues on the range of the parameters you
: can choose.
:
: ,

L*******t
发帖数: 2385
18
true,这个我也考虑过,任意选一个SDF是为了calibration方便,并不是最好的办法。
我记得有些个paper讨论过怎么选择所谓的RN measure。但是我不知道有谁比较过不同
的EMM下,pricing的优劣的。虽然Incompleteness的情况下,He and Pearson(1993)
已经吧所有的EMM都characterize出来了。
现在有一些人做Recovery theorems,从option price information去recover SDF,
RN的underlying distribution。我觉得这是个很好的方向,至少对modeling有一定的
启发
financial market是不是完备的,可能这个还有点小争议。到底是资产多还是Shock多
?shock如果是不可列的话,completeness可能就要breakdown了。这个问题我也想不太
明白。

technique
SDF

【在 C******n 的大作中提到】
: 你不能说因为 "this model"是Incomplete,“所以”你就可以任意选一个SDF。
: 对于你所感兴趣的market,例如option market,fact会告诉你SDF应该长什么样。
: financial market大体上是complete的,虽然有很多的noise,所谓incomplete只是因
: 为你的model focus on 1 asset and 2 or 3 shocks,这只是你的modeling technique
: 的限制,不代表你应该这么反映现实。
: 举个例子,BS + SV或者BS + jump都是incomplete,但merton和heston都是怎么确定SDF
: 的,也就是说他们做了何种假设?是否合理?不合理的话现在大家在干什么?

L*******t
发帖数: 2385
19
这个我觉得并非如此。是否complete这个还有争仪。如果按照这个思路走下去,
completeness也有可能是个stupid idea,有什么证据说明3资产3shock的模型是个好模
型?
其实所有的模型都是错的,咱都在用错误的模型去fit数据,然而市场就一定是正确的
?这么想下去就很悲观,金融和自然科学很不一样。
大家在做的都是在放松一些假设,但是我觉得,得有一些概念上的break through,否
则期权定价现有的方法会是一个死胡同

【在 C******n 的大作中提到】
: The actual market并没有模型里所说的incompleteness。incomplete的是stupid
: model。换句话说,这是个模型的缺陷。我们需要做假设来弥补。
: actual market上有Jump和SV吗?市场还在吗?
:
: do

L*******t
发帖数: 2385
20
ok,market is incomplete
但是任意shock都写一张contract似乎也是很困难的,因为有时候你根本不知道shock到
底是啥。
model是incomplete的话,SDF有很多,到底哪一个正确,似乎还没有定论,但是从
calibration的层面上说,我可以选一个使得模型在PQ下面的参数差不多,这个是俺的
意思啦~~
我觉得研究一下不同SDF给出的价格区间是一个很好玩的问题,而且不同SDF对cross
sectional data fit的情况应该也有不同吧。我好像没读过这方面的文章。
incompleteness给现在不那么完美的模型一点希望。
但是如果你要complete你的model总是有办法的,比如general equilibrium的时候,你
总可以加一些zero net supply的asset,让市场complete,但是complete market下的
puzzle太多。。

【在 C******n 的大作中提到】
: market肯定是incomplete的,没人能买卖任意数量。
: 在SV model里是incomplete,但这是this model的缺陷,因为实际上有比该模型更多的
: product存在但没考虑。Potentially,任意的shock都可以write contract。正如你所
: 说,vol是个market,而且liquidity足够好,for the purpose of SV modeling,可以
: 算是complete。
: 我主要说的是上面童鞋想模型有点太literal,model说incomplete就能任意设置SDF,
: 这是不合理的。
:
: to
: a

相关主题
Pricing a trinomial call option关于选Offer。。。大家来给看看
还是分红和期权的问题问个绿书上的问题 - delta hedging
再借人气问一个关于期权Cross Sectional Data的问题【Option questions】if you have a model to give better vol smirk and term structure
进入Quant版参与讨论
L*******t
发帖数: 2385
21
true and not true
SDF距离能price option dynamics and cross sectional data的情况还差很远
一个很容易的想法是,市场总能pickup一个SDF,但是SDF这条路是否是正确的?
这个似乎很难justify

【在 p******i 的大作中提到】
: exactly,
: incomplete market means THE MARKET will choose one of the equivalent RN
: measure not you
:
: you

L*******t
发帖数: 2385
22
我记得读过一篇paper,Jumpe Heston
用历史数据和option data共同calibrate。
但是更复杂一些的model,用optiondata就有些困难了吧。。

【在 q*****l 的大作中提到】
: 请问一般来说SV model都是用option的market price来做calibration的,然而同样的我
: 们貌似也可以用market price of stock来直接估计SV model的参数。
: 请问这两种方法哪个更普遍一点?

s******e
发帖数: 1751
23
why do u need sv and what instruments will have exposure to sv parameter?
does sv model give u a better delta hedge result?

【在 L*******t 的大作中提到】
: 我记得读过一篇paper,Jumpe Heston
: 用历史数据和option data共同calibrate。
: 但是更复杂一些的model,用optiondata就有些困难了吧。。

L*******t
发帖数: 2385
24
so instead of sv do we have any other good alternatives?......

【在 s******e 的大作中提到】
: why do u need sv and what instruments will have exposure to sv parameter?
: does sv model give u a better delta hedge result?

L*******t
发帖数: 2385
25
我觉得那些搞modeling的童鞋们有些偏执的地方就在于,明知道model是不完美的,偏
想要找到一条接近完美的模型。。问个比较practical的问题,也是出于我的无知,BS
的delta hedging,效果好么?。。。。。

to
a
big

【在 s******e 的大作中提到】
: market could be incomplete. (i want to trade bitcoin calls. who is going to
: make a market for me?)
: given that, what's the point of talking about those fancy model (there is a
: point for some products, but that's very nichy, to which most of noices do
: not have exposure)?
: there are jump and sv on the market (credit risk and VIX vol). both are big
: market.

s******e
发帖数: 1751
26
almost all models are designed for risk management.
you need sv model if u have sv exposure. in equity world sv exposure is
small after 2008.

【在 x********o 的大作中提到】
: do you forgot about risk management? in many cases you have to have SV
: models, or even SV + jumps
:
: to
: a
: big

L*******t
发帖数: 2385
27
我觉得你是比较懂options的。。。
想问你一个问题。。在industry,大家是怎么看待deep OTM options的价格的?
出了用BS算一个IMPLVOL以外,我还不知道有什么SV模型能算准他们的价格的,
Local SV的话,都在用Gyongy theorem,这个定理只保证marginaldistribution是相同
的。。

【在 s******e 的大作中提到】
: almost all models are designed for risk management.
: you need sv model if u have sv exposure. in equity world sv exposure is
: small after 2008.

C******n
发帖数: 9204
28
大家都是在通往真理的路上。搞点breakthrough都是以10年为单位计算的,病态的创造
力。
从equity和fixed-income (US treasury, corp bond) 来看,general lesson是2-3个
shock可以pick up 大概90%的波动。但如果你impose一个param model,三个就不够了
。 3-factor affine/quad TS model表现很糟糕。但这不完全是个数的问题。
所以在我看来completeness问题只有理论意义,没有实际意义。由于人是很复杂的,所
以潜在的,你就是搞17个因素都不够,但empirically主要因素也没几个。
recovery thm那个定理让我震精了。但用起来还是需要一些假设,并没有非常perfect
。但这个想法很nb,把之前的人都日了。
SDF encode了人的relative valuation。按照consumption-based的说法,应该是
marginal util的ratio(recovery them和Heston都是这么用的)。general lesson是人
们非常disaster-aversion(除了recovery thm一文),但这点还没有共识。

BS

【在 L*******t 的大作中提到】
: 我觉得那些搞modeling的童鞋们有些偏执的地方就在于,明知道model是不完美的,偏
: 想要找到一条接近完美的模型。。问个比较practical的问题,也是出于我的无知,BS
: 的delta hedging,效果好么?。。。。。
:
: to
: a
: big

s******e
发帖数: 1751
29
have you backtested holding an 105% OTM SPX call? I don't see anything wrong
w/ just using BS vol. In another word, i don't see any needs to invent some
fancy model for an deep OTM european option.
before you spend so much time reading those published stuff, what problem
you try to solve? Recall why heston invented stochastic vol? what problem
he was trying to solve?

【在 L*******t 的大作中提到】
: 我觉得你是比较懂options的。。。
: 想问你一个问题。。在industry,大家是怎么看待deep OTM options的价格的?
: 出了用BS算一个IMPLVOL以外,我还不知道有什么SV模型能算准他们的价格的,
: Local SV的话,都在用Gyongy theorem,这个定理只保证marginaldistribution是相同
: 的。。

c**********6
发帖数: 18
30
Remember it;s non-arbitrage argument makes option prices meaningful and BS a
noble prize. If the price is for trading purpose and u can't reasonably
replicate the prices by hedging, what does the option price mean and really
matter to u? I think you can't discuss option pricing without hedging. No
hedging is perfect but some hedging technique is meaningful for certain
models. By reversing this thinking u can know that the deep otm option are
priced that way are because market participants have reasonable hedging tech
that can arbitrage around it. I don;t believe anyone can accurately dynamic
hedge it,but the prices can;t be too off.
In this sense, u may not want to go to theoretically incomplete pricing
model and ignore the fact that prices in the market are not from models but
from supply/demand like any other traded assets.

【在 L*******t 的大作中提到】
: 我觉得你是比较懂options的。。。
: 想问你一个问题。。在industry,大家是怎么看待deep OTM options的价格的?
: 出了用BS算一个IMPLVOL以外,我还不知道有什么SV模型能算准他们的价格的,
: Local SV的话,都在用Gyongy theorem,这个定理只保证marginaldistribution是相同
: 的。。

相关主题
Derivative Quant离开美国去哪好?Yahoo Finance数据源一问
问个spread预测问题[合集] 面试问题请教
请问heston model怎么hedge?战五渣再问个implied volatility的问题
进入Quant版参与讨论
s******e
发帖数: 1751
31
hedging OTM option is an easy job. don't see ppl complaining about it.
supply/demand, rather than models, governs market.

a
really
tech
dynamic

【在 c**********6 的大作中提到】
: Remember it;s non-arbitrage argument makes option prices meaningful and BS a
: noble prize. If the price is for trading purpose and u can't reasonably
: replicate the prices by hedging, what does the option price mean and really
: matter to u? I think you can't discuss option pricing without hedging. No
: hedging is perfect but some hedging technique is meaningful for certain
: models. By reversing this thinking u can know that the deep otm option are
: priced that way are because market participants have reasonable hedging tech
: that can arbitrage around it. I don;t believe anyone can accurately dynamic
: hedge it,but the prices can;t be too off.
: In this sense, u may not want to go to theoretically incomplete pricing

c**********6
发帖数: 18
32
Surely OTM optioin are much easier to hedge than deep ITM options. I think
Leinhardt is taking one step back for argument sake. The universal hedging
error still exist.(bid-ask spread, hedging frequency, model u use, hedging
ratio parameters). But the error is unavoidable and can be absolutely
sustained within a tolerable range.
I don't think this kind of error can be eliminated thus reduce the bid-ask
spread in the option market unless there is another powerful non-arbitrage
argument emerging in academia which is gonna be a breakthrough which may be
very unlikely to happen. However I don't think stoch vol is the way to go,
if now we are talking academia.

【在 s******e 的大作中提到】
: hedging OTM option is an easy job. don't see ppl complaining about it.
: supply/demand, rather than models, governs market.
:
: a
: really
: tech
: dynamic

L*******t
发帖数: 2385
33
Oh,我以前从来没有做过更加practical一点的事情。包括你说的backtesting(hedging
?)
如何构造Impl Vol曲面,如何用这个加上BS进行定价,Local Vol的pros and cons都是
从文章里面学到的。事实是我也没有机会接触这些实际的操作。
theoretical world里面模型的spirits是找到一个模型,能够对vol simle 及其
dynamics fit得较好。而且模型参数还必须保持尽量不变,并且initial values, e.g.
stovol, etc也必须不变。这个看似苛刻的要求就是我奋斗的目标,但是我感觉这条路
不好走。

wrong
some

【在 s******e 的大作中提到】
: have you backtested holding an 105% OTM SPX call? I don't see anything wrong
: w/ just using BS vol. In another word, i don't see any needs to invent some
: fancy model for an deep OTM european option.
: before you spend so much time reading those published stuff, what problem
: you try to solve? Recall why heston invented stochastic vol? what problem
: he was trying to solve?

L*******t
发帖数: 2385
34
glad to see so many experts here!
很多academic paper都似乎只是在讨论价格,虽然也有许多直接讨论hedging的。但我
的重点似乎都放在了前者,业界似乎更加在乎hedging吧。因为价格总有办法得到。而
hedging看上去tricky一些。。
在option market,应该有arbitrage的机会吧?或者因为流动性,没人去地上捡这些小
钱?
我的感觉是no arbitrage-> posotive sdf-> change of mmeasure这套东西很难
reconcile
market data了,local stochastic vol这套工具只能对path independent的产品定价
,似乎也走不通了。我曾经向我老板建议不通的产品似乎隐含不同的sdf,因为人们交
易不同产品的时候对future的views是不一样的,被无情的批了一顿。
对option pricing和hedging,你有什么建议么?

be

【在 c**********6 的大作中提到】
: Surely OTM optioin are much easier to hedge than deep ITM options. I think
: Leinhardt is taking one step back for argument sake. The universal hedging
: error still exist.(bid-ask spread, hedging frequency, model u use, hedging
: ratio parameters). But the error is unavoidable and can be absolutely
: sustained within a tolerable range.
: I don't think this kind of error can be eliminated thus reduce the bid-ask
: spread in the option market unless there is another powerful non-arbitrage
: argument emerging in academia which is gonna be a breakthrough which may be
: very unlikely to happen. However I don't think stoch vol is the way to go,
: if now we are talking academia.

L*******t
发帖数: 2385
35
有一个我想弄清不过可能永远也弄不清的是,交易的人到底用神马模型?

a
really
tech
dynamic

【在 c**********6 的大作中提到】
: Remember it;s non-arbitrage argument makes option prices meaningful and BS a
: noble prize. If the price is for trading purpose and u can't reasonably
: replicate the prices by hedging, what does the option price mean and really
: matter to u? I think you can't discuss option pricing without hedging. No
: hedging is perfect but some hedging technique is meaningful for certain
: models. By reversing this thinking u can know that the deep otm option are
: priced that way are because market participants have reasonable hedging tech
: that can arbitrage around it. I don;t believe anyone can accurately dynamic
: hedge it,but the prices can;t be too off.
: In this sense, u may not want to go to theoretically incomplete pricing

c**********6
发帖数: 18
36
Sry.. this is gonna involve some proprietary stuff here. I may be not able
to talk about it here :) Saw ur previous post. now i am clear about what u
trying to achieve here. Guess u r struggling with vol surface.There are only
a few popular vol surface modes industry like to use as a basic (surely not
local vol model) and u can search and start from there :)

【在 L*******t 的大作中提到】
: glad to see so many experts here!
: 很多academic paper都似乎只是在讨论价格,虽然也有许多直接讨论hedging的。但我
: 的重点似乎都放在了前者,业界似乎更加在乎hedging吧。因为价格总有办法得到。而
: hedging看上去tricky一些。。
: 在option market,应该有arbitrage的机会吧?或者因为流动性,没人去地上捡这些小
: 钱?
: 我的感觉是no arbitrage-> posotive sdf-> change of mmeasure这套东西很难
: reconcile
: market data了,local stochastic vol这套工具只能对path independent的产品定价
: ,似乎也走不通了。我曾经向我老板建议不通的产品似乎隐含不同的sdf,因为人们交

C******n
发帖数: 9204
37
non-exotic用BS。exotic用SV等复杂模型。在报价的时候可以再有些调整好挣点钱。
hedging的话,有auto hedging,或者靠trader自己判断,make a bet,做个partial
hedging。

【在 L*******t 的大作中提到】
: 有一个我想弄清不过可能永远也弄不清的是,交易的人到底用神马模型?
:
: a
: really
: tech
: dynamic

1 (共1页)
进入Quant版参与讨论
相关主题
问个绿书上的问题 - delta hedging问个volatility surface的问题
【Option questions】if you have a model to give better vol smirk and term structureHeston model calibration
Derivative Quant离开美国去哪好?暑期实习求推荐。。
问个spread预测问题求教Bessel过程
请问heston model怎么hedge?MFE deadline的问题
Yahoo Finance数据源一问弱问一题,关于binomial的
[合集] 面试问题请教成绩单问题-请问CMU在读同学,或今年的申请人
战五渣再问个implied volatility的问题Pricing a trinomial call option
相关话题的讨论汇总
话题: sv话题: model话题: market话题: hedging话题: sdf