q*d
发帖数: 22178 | 1 下到这本书的电子版无数天了,每次读都被书最开始的option price,
Black-Scholes model这些我根本不知所云的概念吓住,没继续下去,
直到近日,版上有人重新提到这本书,终于捏着鼻子把序言部分读完,
看到第一章他怎么学物理的,里面提及一些物理学史的部分,错误无数:
1.这厮是南非人,在开普敦读的本科,1966年到哥大读PHD
2.他的PHD读了7年,正如本版已经有人指出的,他的解释是他在
开普敦没有受到很好的教育,在哥大必须要上3还是4年的课.
3.第一章23页,当提及Fermi的时候,他说"Amazingly,he was also
the theorist who, in the 1930s, had predicted the existence
of the neutrino"--neutrino是Pauli最先预言的吧,名字是Fermi起的.
4.第一章24页,"Willis Lamb and Polykarp Kush,both at Columbia in
the late 1940s,had carefully and accurately mea... 阅读全帖 |
|
g*********d
发帖数: 233 | 2 if your background is in physics,
and knows stochastic simulation,
it takes you only a few weeks to understand
option price models like Black-Scholes. |
|
s*******e
发帖数: 60 | 3 商学院的小女孩by H2O
这是秋季学期的最后一天，外面正在下着雪，天也渐渐黑了下来。教室里暖和极了，热
得小女孩脱掉了外套。她一边想着发信去抱怨空调开得太热，一边把手放在相对凉快的
玻璃窗上。在这又冷又黑的晚上，同学们都回家去了，一个穿短裙的小女孩在教室里坐
着。她从家里出来的时候还穿着一件外套，但是教室里太热了———她现在只穿着一条
吊带裙，紧紧地贴在她美好的身材上。为了凉快点，她把靴子脱掉了，露出一双玉腿。
同学们常常开玩笑说，那双玉腿连凯恩斯看了都会流鼻血。
小女孩一个人做case analysis，我们可以看见她白皙而美丽的脸庞。她面前的讲义是
白白的，没有划过一条线，桌上的草稿纸上只画着一个flow chart。她懒得出去买汉堡
，只是到星巴克买了一杯拿铁咖啡暖手。连executive summary都没写出来，谁也没来
帮她。
可怜的小女孩！她无聊极了，磨磨蹭蹭地开始写SWOT的opportunity/threat部分。白色
的灯把光照在她新烫的发式上，那头发卷曲着披在肩上，用造型发乳精心地打理过，保
持着最初的造型。不过她没注意这些。桌上堆满了草稿纸，教室里飘着一股新设... 阅读全帖 |
|
a***r
发帖数: 594 | 4 from what you said, you have no understanding in the big picture of quant
finance. you do not appear to have studied physics in any depth AT ALL either.
many separate the finance world into buy side and sell side. I happen to
have worked on both sides.
the training a physicist receives makes him/her a very good fit for sell
side quant roles.
In this part of the world, one does not try to forecast the future returns
on financial instruments. Instead one focuses on understanding the behavior
of th... 阅读全帖 |
|
S*******s
发帖数: 13043 | 5 no, Black–Scholes-Merton was awarded nobel until Black died |
|
d****b
发帖数: 24 | 6 一起努力吧。我因为不是好学校毕业,背景也不够强，还是很难。
先自己多准备, 尤其是C++, 先看thinking in C++ , 然后 effective C++等
virtual table, copy constructor, virtual destructor 经常被问.
然后John Hull的那本bible 至少要看熟black-scholes 那块, 要达到能把公式推出来,
立刻写出解的地步, 可能要更多.
上NYU等牛校的FE专业查看课程设置，download 相关的notes,借鉴他们的Resume.
看统计方面的stochastic calculus, time series, hitting time等等. 如果不是数学专
业， 可能需要了解最基本的matrix和finite difference.
觉的差不多再开始正式投resume。一定要来New York, 至少随时可以准备interview.
我从去年12月开始看这些东西, 5月开始正式找, 前几个phone interview 基本几个问题
就结束了. 现在两个月过去了, 边找边看书, 终于开始有些感觉 |
|
B*********h
发帖数: 800 | 7 ☆─────────────────────────────────────☆
dingyb (江南风灵) 于 (Wed Jul 26 00:25:21 2006) 提到:
一起努力吧。我因为不是好学校毕业,背景也不够强，还是很难。
先自己多准备, 尤其是C++, 先看thinking in C++ , 然后 effective C++等
virtual table, copy constructor, virtual destructor 经常被问.
然后John Hull的那本bible 至少要看熟black-scholes 那块, 要达到能把公式推出来,
立刻写出解的地步, 可能要更多.
上NYU等牛校的FE专业查看课程设置，download 相关的notes,借鉴他们的Resume.
看统计方面的stochastic calculus, time series, hitting time等等. 如果不是数学专
业， 可能需要了解最基本的matrix和finite difference.
觉的差不多再开始正式投resume。一定要来New York, 至少随时可以准备i |
|
s*********7
发帖数: 4 | 8 应征Quant。
连面试都谈不上，只是和猎头面谈。
本人背景，国内某大学信号处理专业。来美国从事跟无关金融信号处理研究（postdoc）
。对Quant很感兴趣，也按照板上诸位建议看John Hull的书。black－scholes model倒
没觉得什么，因以前学过随机过程。
今天面谈，小猎头对我挺感兴趣，但是强调financial知识和C++的重要性。大猎头（老
板）姗姗来迟，他是俄罗斯的数学博士，一坐下来就跟我说，“Many people think Qu
ant is hungry for physics and signal processing guys, they are wrong.” 然后
强调Quant要求的是one man work shop,就是一个人设计模型，编程实现，最终交易的能
力，所以建模编程能力一定要又快又强。然后说到black model，“for so many years
, people have been beating the shit out of the Black model”,所以光看看书是m
eaningless，一定要快速实现 |
|
r*****t
发帖数: 286 | 9 ☆─────────────────────────────────────☆
thinkme (thinkme) 于 (Wed Feb 7 22:03:20 2007) 提到:
In Black-Schole's formula, the rate of drift of the stock does not
matter. This is rather counter-intuitive. If I'm a client and I
insist that the option should have higher price with higher drifting
rate, how would you explain it to me?
☆─────────────────────────────────────☆
cuyang (cuyang) 于 (Wed Feb 7 22:20:08 2007) 提到:
risk－neutral
☆─────────────────────────────────────☆
thinkme (thinkm |
|
r*****t
发帖数: 286 | 10 ☆─────────────────────────────────────☆
scotthuang (scott) 于 (Mon Mar 26 13:59:12 2007) 提到:
Suppose in a black-scholes world, r=0. And you have a digital option which
pays 1 dollar at T if S_T > K. If price of the stock doesn't change for a
period of delta t after the issuance, the price of the option goes up.
However, no hedging strategy using bonds and stocks can produce the rise in
value. (stock price doesn't change. Bond price doesn't change either as r =
0)
Isn't it strange?
☆───────── |
|
r*****t
发帖数: 286 | 11 ☆─────────────────────────────────────☆
yfh (简单略过了) 于 (Sat Jun 2 10:57:04 2007) 提到:
我感觉不行呀
谢谢
☆─────────────────────────────────────☆
scarface (人生犹如一场电影) 于 (Sat Jun 2 11:30:28 2007) 提到:
yeah, u can.
☆─────────────────────────────────────☆
yfh (简单略过了) 于 (Sat Jun 2 12:26:07 2007) 提到:
I am alittle confused:
Does bond price follow a log-normal distribution?
Alternatively, does the volatility of the bond price remain approxmately
constant over a period of time?
☆ |
|
B*********h
发帖数: 800 | 12 ☆─────────────────────────────────────☆
snooner2007 (snooner) 于 (Thu Mar 8 21:35:52 2007) 提到:
应征Quant。
连面试都谈不上，只是和猎头面谈。
本人背景，国内某大学信号处理专业。来美国从事跟无关金融信号处理研究（postdoc）
。对Quant很感兴趣，也按照板上诸位建议看John Hull的书。black－scholes model倒
没觉得什么，因以前学过随机过程。
今天面谈，小猎头对我挺感兴趣，但是强调financial知识和C++的重要性。大猎头（老
板）姗姗来迟，他是俄罗斯的数学博士，一坐下来就跟我说，“Many people think Qu
ant is hungry for physics and signal processing guys, they are wrong.” 然后
强调Quant要求的是one man work shop,就是一个人设计模型，编程实现，最终交易的能
力，所以建模编程能力一定要又快又强。然后说到black model，“for s |
|
B*********h
发帖数: 800 | 13 ☆─────────────────────────────────────☆
huaniuniu (huaniuniu) 于 (Sat Mar 24 22:15:57 2007) 提到:
0.0 First steps -- General:
A. Black Scholes and Beyond: Option Pricing Models, N A Chriss
B. Derivative Securities, R Jarrow, S Turnbull
C. Introduction to Mathematical Finance: Discrete Time Models, S R Pliska （
听说经济科学社出了中译本，但是没看到过。）
0.1 First steps -- Interest rates:
A. Fixed Income Analytics, K Garbade
0.3 First steps -- Stochastic Calculus:
A. An Introduction to the Mathematics of Financial D |
|
B*********h
发帖数: 800 | 14 ☆─────────────────────────────────────☆
lunasol (lunasol) 于 (Thu Jul 5 23:02:46 2007) 提到:
想学金融方面的。本科chemical engineering快毕业了。找个manufacturing工作，工
作几年，拿个MBA,进入金融，浪费时间吗？
还是咬牙那个MFE?(一点金融背景没有，programming不强）
希望前辈执教
☆─────────────────────────────────────☆
netghost (Up to Isomorphism) 于 (Thu Jul 5 23:07:39 2007) 提到:
quant, highly recommended.
☆─────────────────────────────────────☆
ThatYear (那年) 于 (Thu Jul 5 23:10:17 2007) 提到:
transport phenomena 学的怎样?
black scholes model也就一 one dime |
|
s**7
发帖数: 4 | 15 Future is the derivative of underlying asset price.
For commodity, asset price is spot price.
Black-schole's stock price is assumed to be random walk.
I figure spot price should be random walk too.
option price is derived based on 'no arbitrage' too. Future price is just
not calculated from stochastic PDE. But I am sure future price, same as
option price, must have volatility.
right? |
|
s**7
发帖数: 4 | 16 When I check Black-scholes option pricing model,
The price of the underlying instrument St follows a geometric Brownian
motion with constant drift μ and volatility σ:
Yes. volatility σ refers to unerlying asset's volatility.
Also check here: http://en.wikipedia.org/wiki/Implied_volatility#Solving_the_inverse_pricing_model_function
the implied volatility of an option contract is the volatility implied by
the market price of the option based on an option pricing model.
////About historical volatil |
|
k****z
发帖数: 550 | 17 自然，整个black-scholes model都是基于markov性。如果不是markov，就有很多
adjustment可以讨论了。有时面试也会碰到这种问题。 |
|
g********e
发帖数: 25 | 18 they don't require me to do any programming.
Just ask me to introduce my resume.
ask about my leadership experience.
asked Black-Schole |
|
z****u
发帖数: 185 | 19 I don't know how to explain. But the truth is if you agree the pdf of the
stock level at a future time has a fat tail, then implied vol cannot be only
a function of strike. Because this will lead to strike-dimension arbitrage
opportunities, which mean local variance will be negative.
On the other hand, you can insist on no-fat-tail and have a consistent
theoretical picture, i.e., the Black-Scholes flat skew. In this case, the
implied vol is a constant. This does not cause any problem since now t |
|
b***k
发帖数: 2673 | 20 ☆─────────────────────────────────────☆
hippophip (hippophip) 于 (Sat Jan 26 18:49:21 2008) 提到:
有没有学的实际意义，现在读MASTER，不知道该不该选这门课。以后找金融的工作，这
些学了可以做什么？会学到：Brownian Motion, Ito integral,Ito's lemma, Black-
Scholes equation, Risk-Neutral Pricing, Partial Differential Equations,
Exotic Options,Interest Rate Models，Valuation of mortgage backed securities
☆─────────────────────────────────────☆
yww (petite) 于 (Sat Jan 26 21:45:38 2008) 提到:
我ft，这和找IT要学OS一样基本属于必修
otherwise只能搞IT里面的网页设计了
securities |
|
b***k
发帖数: 2673 | 21 ☆─────────────────────────────────────☆
blook (布鲁克) 于 (Sat Feb 23 13:32:53 2008) 提到:
我前段好像看到哪本书（严重怀疑是Baxter&Rennie)中讲，
HJM模型在interest rate中的地位和black scholes在option中的地方相当。
可是我发现Shreve和Hull的书中其实并没有这么强调HJM，
所以我严重怀疑以上那个statement的可信度，
可是回去查了半天，也没有找到这句话。
any comment from you guys?
☆─────────────────────────────────────☆
Jadeson (Jadeson) 于 (Sat Feb 23 14:03:31 2008) 提到:
很重要
☆─────────────────────────────────────☆
helmertblock (fun) 于 (Sat Feb 23 14:08:56 2008) 提到:
re
☆────────────────── |
|
J*****n
发帖数: 4859 | 22 不是那么简单的，要看具体tax政策的。
曾经看过一篇文章。说英国政府突然改变derivative得税收政策，结果几个银行亏损无
数。 |
|
c*******d
发帖数: 72 | 23 最简单的tax 政策, interest 和dividend 收税a%, capital gain/loss都不收税 |
|
h**********k
发帖数: 168 | 24 not easy.. there are some papers on this... |
|
T*C
发帖数: 14 | 25 That's a good topic these days. you can write a Phd thesis on it. For short
term traders, that's nothing. for long term fund, big time! |
|
b***k
发帖数: 2673 | 26 ☆─────────────────────────────────────☆
trenchant (N/A) 于 (Thu Sep 20 22:05:18 2007) 提到:
如果 volatility \Sigma 变得无穷大
我算得 d1 正无穷大
d2 负无穷大
那样的话， c = S0; p=K exp(-rT)
看起来很奇怪，是我哪里算错了？
谢谢
☆─────────────────────────────────────☆
Diadora (枯藤老树烤鸭 小桥流水酒家) 于 (Thu Sep 20 22:19:11 2007) 提到:
没错
收益和买股票一样啊
☆─────────────────────────────────────☆
trenchant (N/A) 于 (Thu Sep 20 22:25:24 2007) 提到:
又一个问题想不通
既然 d2 是负无穷大
那么CALL里面的第二项
N（d2）is the probability that the option will be exercised
＝0
也就是说CA |
|
l*****a
发帖数: 119 | 27 选课问题， 麻烦各位高人帮看看应该选那些。谢谢。
1. Mathematics of Finance
Binomial model of stocks, stochastic calculus, martingales and arbitrage,
Black-Scholes equation and pricing derivative securities, interest rate
modeling, introduction to portfolio risk management.
2. Applied Computational Finance
Overview of equity, fixed income and FX markets; summary of continuous time
financial modelling; pricing of vanilla and exotic derivatives; lattice
methods: binomial trees; Monte Carlo methods; numerical methods for |
|
x**p
发帖数: 105 | 28 Assuming V1 and V2 follow two different BMs, use V2 as numerarire to get rid
of one BM and convert that problem to V2*max(0,V1/V2-1), where V1/V2 follow
one dimensional BM. That is a standard B-S problem. You need to consider
correlation between V1 and V2 to get the correct BM of V1/V2. |
|
t**********a
发帖数: 166 | 29 One addition: if V2 has dividend, you need to take that into account for
numeraire.
check derivation of Margrabe formula
rid
follow |
|
|
c******e
发帖数: 25 | 31 stike=0, model spread dynamics? |
|
j*****o
发帖数: 6 | 32 If you said VaR is useless clearly you've never really worked in the market
risk business. In most cases, it is still the single most important measure
people looking at when speaking about market risk. A strong evidence is how
regulators still place great emphasis in VaR.
That said, one is easy to see that VaR is far from perfect theoretically.
However, no models even the Black-Scholes are perfect in the finance world.
People use them because emperically they give satisfactory results.
In risk |
|
h*****u
发帖数: 38 | 33 You mean black scholes equation or formula? |
|
r**u
发帖数: 69 | 34
i'm assuming these 2 assets are stocks with lognormal distributions.
K
this is a spread option. don't think there is closed formation solution.
this is not too hard as S1*S2 is lognormal itself. this is standard black-
scholes. |
|
h*******n
发帖数: 24 | 35 1. strongly agree with considering S1*S2 as a standard black-scholes.
2. for S1-S2, the "lognormal" way may not work... does that make sense if i
calculate all deltas and do "delta hedge"? |
|
p****a
发帖数: 631 | 36 bond: d(\beta_t) = r\beta_t dt
stock A: dSt = aStdt+bStdBt
stock B: dPt = cPtdt+dPtdWt
where Bt and Wt are both BMs, but independent to each other.
Question: Suppose you have one share of stock A. Then how to price an op
tion that gives you a right, but not the obligation, to exchange one sha
re of stock A to one share of stock B at time T?
Any clue? Thanks! |
|
q********u
发帖数: 53 | 37 Exchange option.
exp(-rt)max(S-P,0)=Pexp(-rt)max(S/P-1,0)
Then std BS with adjust term Pexp(-rt).
Using Girsanov Theorem, the likelihood will be canceled by the expansion of
P. |
|
p****a
发帖数: 631 | 38 thanks, I'll try it.
of |
|
b***k
发帖数: 2673 | 39 ☆─────────────────────────────────────☆
potala (天涯宝贝) 于 (Thu Dec 11 22:11:23 2008) 提到:
dSt = u St dt + sigma St dWt
dBt = r Bt dt
如果u 跟sigma都是常数呢，下面的关系都成立，
哪如果u跟sigma 不是常数，而是u=u(St), sigma=sigma(St)的话，
下面这些关系还成立么，为什么？
1） C(S,t,K,T) = C(S,0,K,T-t)
2) C(aS,t,aK,T) = aC(S,t,K,T)
3) C(S,t,K,T) is an increasing function of S
4) C(S,t,K,T) is a decreasing function of 0
5) C(S,t,K,T) -> max(S-K,0) as t->T
6) 0<=dC/dS<=1
7) if S>K, then dC/dS->1 as t->T
8) if S0 as t->T
☆ |
|
j******i
发帖数: 6 | 40 To anybody who knows black-scholes model and stochastic differential
equation or anybody who knows control theory.
dS=αSdt+βSdB, S denote the price of a stock depends on t,S(t), α is the
drift or rate of return, β is the volatility, B is brownian motion.
How do I incorporate the control parameter γ , and a number “ L “defined
as dN/dS, where N denotes the amount of asset traded and S denote the asset
price so that dS=(αS+γ/L)dt+βSdB
defines a differential equation.
Thank you in advance |
|
G*****u
发帖数: 1222 | 41 面试的人是个senior portfolio analyst 10年前从CMU MSQF毕业
应该是fixed income analysis方面的东西
面试的内容会是怎么样呢？
我邮件里问了面试官 他说This is an exploratory interview to try to determine
the fit of your skill set with the position I am trying to fill.
好像是句废话
我需要看看black scholes那些东西吗？这种职位的常用软件是什么？
大家指教下 明天早上就面了 以前从来面过类似职位 谢谢
update:
开始部分：
面试官介绍了下这个职位
编程部分：
最开始是问了编程经验
很详细的问了SAS,MATLAB,VBA,splus的区别
还问了SQL和ACCESS的优缺点
统计部分：
再往下问了我学过的统计课程
让我仔细说了时间序列分析 问了学过的时间序列分析model
然后给了个实例 问如何做这个时间序列分析
问了population 和 sample 的区别 以及如何去除sample bias
接 |
|
h*******a
发帖数: 41 | 42 A straddle = long a put and a call at the same time. We know that when
interest rate goes up, the value of future cash flow decreases, therefore
the value of a call goes up, while that of a put goes down.
Question: does the value of a straddle goes up or down when interest rate
goes up? Give intuitive arguments instead of Black Schole formula. |
|
k**8
发帖数: 14 | 43
Thank you, NYUTT. Can you explain more about how to use libor for implied
volatility calculation? My confusion is that libor rates are not continuous-
compound rates. How can I plug libor into black scholes? Do I need to
convert them first?
Thanks, again. |
|
z****u
发帖数: 185 | 44 "但是后面讲BSM的时候说BSM只能用于Eu option pricing，难道BSM不能用于Am call
option pricing吗？"
My understanding is BSM(Black-Scholes Model) works for both, but the boundary conditions for Euro and Am are different.
stock settlement和future settlement: shouldn't differ too much. |
|
s*****r
发帖数: 10 | 45 f_{t} + r*s* f_{s} + sigma^2*s^2 * f_{ss} - r*f = 0
一般是用Crank-Nicolson是2阶的误差，如果我想提高到三阶以上的话，有什么算法比
较好
的？
主要是做出一个算法以后是不是stable的很难看出来 |
|
s*******s
发帖数: 1568 | 46 rk3 plus standard fourth order discretization in spatial |
|
J*****n
发帖数: 4859 | 47
Richardson extrapolation? |
|
|
o******e
发帖数: 3522 | 49 crank-nicolson是unconditionally stable的，不用看，肯定converge.你是说误差大
小吧？ |
|
l******i
发帖数: 1404 | 50 二阶精度一般就够了，多出的自由度不如拿去提高别的部分，例如dissipativity。看
你算法矩阵的eigenvalue是不是小于等于1去决定是不是unconditionally stable。 |
|
