c******r
发帖数: 300 | 1 E(X_t)=0 holds since we can exchange the integration with taking expectation.
Your answer E(X_tX_s) is also correct as well, I verified it in another way. |
|
J*******g
发帖数: 267 | 2 chopinor, could you plz post your derivation, coz I am just curious to see
other way of handling this type of integrals. Thanks a lot.
expectation.
way. |
|
p****a
发帖数: 631 | 3 can you elaborate how you verify it in another way?
thanks!
expectation.
way. |
|
|
j*******d
发帖数: 55 | 5 ratio of an expected value and an std. deviation, 就是： mu/sigma, 数学上有
什么术语表达吗？我只知道skewness 和kurtosis，不知道这个一次项的叫什么，或者
根本就没有这么个术语？谢谢。 |
|
k**k
发帖数: 61 | 6 matlab感觉更实用也更方便些. 这个问题也看到过一些解法，比如设objective
function为:
a + max(- portfolioReturn - a, 0)/(0.05 * number of sample)
其中a是一个适当的初始值, 对于每一个weight scenario， portfolioReturn就是对应
的10000个simulated portfolio returns (从之前的10000个correlated stock return
里算出）. 0.05代表的是5%的confidence interval下的expected shortfall
由此直接用fmincon就能找出最后的优化weight。对此不怎么理解，但试验了一下这个
确实是给出了正确解，不知道有没有懂行的能comment一下？或者建议一个更合适的版
问问？ |
|
k**k
发帖数: 61 | 7 从概念上讲，这个优化问题没什么难的．但我想用matlab中现有的功能去实现，比如说
最常用的constrained optimization中的fmincon函数．难点就是如何设objective
function.要能做到在每一个weight scenario下能自动找出对应的expected shortfall
．之前我提的objective function就是一个workable的例子，只是我没看懂．谁能有具
体的idea? |
|
s***e
发帖数: 267 | 8 Interesting...
If you assume a is the correct 95% Var, then you get the right weights which
minimizes the expected shortfall. Now if you have the right weights, and
just try to vary "a", you should get the right a, i.e. the 95% var.
So it turns out that a* and w* (the optimal var and optimal weights) are
solution of the optimization problem.
you
returns |
|
r*g
发帖数: 3159 | 9 似乎笔误，max前少一个求和，然后就是 - expected shortfall 定义。
return |
|
M****i
发帖数: 58 | 10 Another method:
To compute EcosB(t), Ito's formula gives
cosB(t)=1-\int_0^t sinB(s) dB(s)-(1/2)*\int_0^t cosB(s) ds
Taking expectation,
EcosB(t)=1-(1/2)*\int_0^t EcosB(s) ds.
So that EcosB(t) is the unique solution of ODE:
df(t)/dt=-(1/2)f(t), f(0)=1.
Hence EcosB(t)=exp(-t/2).
The same method yields that E(cosB(t))^2 is the unique solution of ODE:
df(t)/dt=1-2f(t), f(0)=1. So that E(cosB(t))^2=(1+exp(-2t))/2 and hence
var(EcosB(t))=(1+exp(-2t))/2 -exp(-t)=(1-exp(-t))^2/2. |
|
a****3
发帖数: 19 | 11 问expectation of salary Top Bank Investment Management Division. Associate
level in New York. 有经验的人士指点一下。 |
|
l***m
发帖数: 920 | 12 I think 150k is rather the pay of last century.
With tighter regulations and more labor supply, one should expect to get
paid less within banks. |
|
l*******1
发帖数: 113 | 13 for a n-sided polygon, starting from any vertex, whats the expected time
until all the other vertices are visited, if we randomly go to the two
neighboring vertices with equal probability?
thanks |
|
M*******i
发帖数: 82 | 14 the expected time from state (n-1) to state n is simply n - 1
sum(i-1) i from 1 to n -> n(n-1)/2 |
|
c**s
发帖数: 13 | 15 本人fresh graduate，刚收到一个verbal offer，recruiter说base应该在85k -- 95k
之间，bonus则根据表现。想请教下各位：
1. 现在街上特别是jpm家的package，fresh master应该expect多少啊？85k这个算是个
正常值么？
2. jpm的bonus一般会给多少啊？
3. jpm催得紧，但是我接下来还有gs和aqr的面试，想请教下有了j的这个offer是否还
值得继续面那几家。
跪求坂上各位神人解答。 |
|
M*******a
发帖数: 870 | 16 不是搞option, 也忘了黑鞋子公式了，瞎说下大牛别笑话Option price无伦是根据黑鞋
子公式还是binomial tree啥的都是based on historical data，expected return 是
forward looking, 本身就具不可预测性，option price 如果based on 不可预测的
data，你让人怎么定价？
关？
worth |
|
g*******g
发帖数: 59 | 17 个人一个理解，option需要financing， financing 成本永远都是r_f,和expected
return无关 |
|
p*********s
发帖数: 61 | 18 Because the drift (expected return) is "delta hedged"?
关？
worth |
|
A********i
发帖数: 97 | 19 那好，那什么是expected return? |
|
A********i
发帖数: 97 | 20 Do you mean there is really no expected return when standing at today and
looking ahead, there is only implied volatility?
based |
|
l*******e
发帖数: 17 | 21 我也是这么想的，但是具体来实现有点小瑕疵，比如最大两个是（10,10）的话，另外
两个可以是1到10的任何数，然后是要考虑各种排列。如果最大两个是（10,9）的话，
另外两个数可以是1到9的任何数，然后考虑排列。现在我纠结的就是具体每个情况的排
列怎么算（本人非统计和数学出身，所以功底比较差）。而且这个做法要解决follow
up的问题会比较头疼。假如数据流的长度是8，要算最大4个数的乘积的expected value
的话，用这个办法可能比较麻烦。 |
|
s*********h
发帖数: 648 | 22 2 same envelopes, each has some money. and one envelope has twice the money
as the other one. you choose one envelope and get the money. if you randomly
choose one, and you see it is $100, and people give you a chance to switch.
should i switch?
the 2nd situation is, you randomly choose one, but don't know how much there
is. And people always give you the option to switch. should i switch in
this case?
I think we should not switch in the 2nd situation, or our expectation will
go to infinity that |
|
s******n
发帖数: 95 | 23 Hi all,
For entry level Research Data Coordinator position in biomedical field in
company not school, fresh master, at Maryland, what is the salary
expectation range can I say? Thank you for your advice!! |
|
l**n
发帖数: 40 | 24 【 以下文字转载自 EE 讨论区 】
发信人: lien (lien), 信区: EE
标 题: 问一个expectation计算的问题
发信站: BBS 未名空间站 (Tue May 4 00:55:21 2010, 美东)
如果一个信号的表达式是x[n]=0.1e[n]+1.8x[n-1]-x[n-2]+0.35x[n-3]，其中e[n]是
zero-mean,i.i.d,Gaussian-distributed random process with unit variance。现在
想计算E[x[n]x[n]]和E[x[n]x[n-1]],请问应该是怎么计算的？
谢谢~ |
|
s**u
发帖数: 383 | 25 刚开始学WINBUG, 写了个小程序, model 问题, 但是load data 时说 expected
variable name. 请问哪里出错了. 谢谢.
model {
for (i in 1:9) {n[i] <- nwin1[i] + nwin2[i]
winn1[i] ~dbin(p[i],n[i])
logit(p[i]) <- d[dind1[i]]-d[dind2[i]]}
for ( i in 1:2) {d[i] ~ dnorm(0, 0.001)}
for ( i in 4:5) {d[i] ~ dnorm(0, 0.001)}
d[3] <-0
}
list (dind1 = c(1,1,1,1,2,2,2,3,4),
dind2 = c(2,3,4,5,3,4,5,5,5),
nwin1 = c(9,12,6,27,9,12,12,2,2),
nwin2 = c(10,2,3,2,5,3,3,0,4))
list(d =c(0, 0, NA,0,0)) |
|
s**u
发帖数: 383 | 26 "check model" and "load data"没问题，一complie就显示
expected multivariate node
那位有经验的讲讲这个是什么错误？
model {
for (i in 1:k-1) {
for (j in i+1:k){
win1[i,j] ~ dbin(p[i,j], n)
logit(p[i,j]) <- d[i]-d[j]}
}
d[1:k] ~dmnorm(dmu[ ], prec.Sigma[,])
## Inverse Wishart Prior for Sigma
dmu[1:k] <-0
prec.Sigma[1:k,1:k] ~dwish(Omega[,],k)
Sigma[1:k,1:k]<-inverse(prec.Sigma[,])
for (i in 1:k){ Omega[i,i] <- 0.001}
}
Data:
... 阅读全帖 |
|
s**u
发帖数: 383 | 27 "check model" and "load data"没问题，一complie就显示
expected multivariate node
那位有经验的讲讲这个是什么错误？
model {
for (i in 1:k-1) {
for (j in i+1:k){
win1[i,j] ~ dbin(p[i,j], n)
logit(p[i,j]) <- d[i]-d[j]}
}
d[1:k] ~dmnorm(dmu[ ], prec.Sigma[,])
## Inverse Wishart Prior for Sigma
dmu[1:k] <-0
prec.Sigma[1:k,1:k] ~dwish(Omega[,],k)
Sigma[1:k,1:k]<-inverse(prec.Sigma[,])
for (i in 1:k){ Omega[i,i] <- 0.001}
}
Data:
... 阅读全帖 |
|
y**u
发帖数: 4 | 28 Dose anyone know how to calculate the expectation of W=X^a/(X+b)^(a+b)? where X
=exp(u+s^2*Z), Z is a standard normal random variable (which implies that X
follows a lognormal distribution). please consider a, b, u, and s as known
constants.
Many Thanks! |
|
c*****m
发帖数: 4817 | 29 phone interview后，hr发了封信问salary expectation，该怎么回答呢？
LA的一个medical center，应该是non-profit，title是sr.statistician
网上查的结果差别很大啊，<50 - >100K都有
本人统计ms，３年工作经验
有什么建议，谢谢 |
|
v******3
发帖数: 16 | 30 马上第二轮面试了，expected salary该说多少呢？大牛们！ |
|
a********g
发帖数: 651 | 31 I am using survexp to calculate expect survival curve in R right now, but
found out that the real people life table in R stopped at yr 1990. after yr
1990, there is no real data for life table.
Have anybody had same issue before? how could I solve it?
Thanks a lot. |
|
k***e
发帖数: 7933 | 32 哈哈， 我老98年就在这个bbs玩过这个expect了。 |
|
c******n
发帖数: 2273 | 33 【 以下文字转载自 Seattle 讨论区 】
发信人: daemonself (mit行为艺术专业博士后导师), 信区: Seattle
标 题: Re: 从99% 脑干恶性肿瘤（expected life one year）到最后确诊只是
发信站: BBS 未名空间站 (Thu Sep 18 12:25:05 2014, 美东)
美国大部分医生到了中国早就被打死好多回了。 |
|
z*******n
发帖数: 1034 | 34 February 18, 2015 12:29 PM
Daniel Terdiman
Two-thirds of Americans think that we’ll be seeing drones dropping off
things like clothes, books, pet food and more within half a decade. And 80
percent of consumers say retailers offering quick drone delivery will be
more likely to get their business.
That’s the conclusion of a new general retail study from Walker Sands
Communications, a Chicago-based PR and social marketing agency.
The full study, titled “Reinventing Retail: What Businesses Need to K... 阅读全帖 |
|
C******g
发帖数: 2930 | 35 With Nintendo once again skipping the event and the lack of interests from
Japan in the Xbox 360, Sony will be heavily spotlighted at this year’s
Tokyo Game Show. In fact, Kaz Hirai is having a special keynote at the
convention. So what should fans expect from Sony at Tokyo Game Show? Let’s
take a look at a few possibilities:
Gran Turismo 5
After being in development for years, it looks like Gran Turismo 5 may
finally be coming out relatively soon. With a 2009 release date a
possibility, you can |
|
w**t
发帖数: 1068 | 36 http://supersuxing.blog.sohu.com/106914760.html
实在懒得贴图，无数张帅图，大家点link自己看～
2008-12-20 | Milestone Expected
标签： 苏醒 allen
MC Su
16号晚上第一次作为主持人参加了2008时尚先生颁奖典礼 嘉宾包括很多艺人和导演还
有时尚界人士 从红地毯到场内前后差不多3个小时的晚宴活动中感受体会了很多带有时
尚元素的东西也了解了一些时尚资讯同时得到了很多主持工作的经验 算起来这是第一
次做大型活动的现场主持 多谢翔哥妮姐和柳岩姐的帮助 尤其是柳美女作为我的
partner从头至尾都很照顾我也教会我很多现场主持的经验 谢谢^^ live show毕竟和录
制节目不同 以往没有台本的情况下反而比较容易放得开思维也可以天马行空因为允许
NG出错了可以重新录过或者在后期剪掉 可是live要求最重要的是不能出错 哪些细节要
强调说明哪些方面要忌讳提及都必须很注意 然后再看怎样把控好整个节奏和气氛 如果
再加上幽默的言语精彩的对白才可以达到游刃有余的程度 所以MC其实是一个非常有难
度和挑战 |
|
|
T*********s
发帖数: 17839 | 38 The exchanges will start putting the trading breaks into effect as early as
Friday for six months. The New York Stock Exchange will begin Friday's
trading session with five stocks: EOG Resources Inc., Genuine Parts Co.,
Harley Davidson Inc., Ryder System Inc. and Zimmer Holdings Inc. The
exchange will gradually add other stocks early next week, expecting to reach
by Wednesday the full number that will be covered.
The plan for the "circuit breakers" was worked out by the SEC and the major
exchang |
|
G*********8
发帖数: 644 | 39 ADP is better than expect. |
|
G*********8
发帖数: 644 | 40 ADP is better than expect. |
|
m*z
发帖数: 2356 | 41 Google missed Wall Street’s expectations for the bottom line. The search
giant notched non-GAAP EPS of $6.45, a bit shy of the $6.52 consensus.
Revenues, excluding those shared with partner websites, were $5.09 billion,
better than the $4.98 billion consensus. After hours, the shares are down
roughly 4%. |
|
K********g
发帖数: 9389 | 42 实际上是这个，并不差，WSJ蓄意打压。
Speaking of earnings, Bank of America Corp. ( BAC ) said that its second-
quarter profit fell to $3.1 billion, or 27 cents per share. Net income
applicable to common shareholders was $2.8 billion, while total revenue net
of interest expense was $29.2 billion. Analysts had expected earnings of 22
cents per share on revenue of $29.6 billion. BAC shares were last seen lower
by 0.84% in pre-market trading.
Read more: http://community.nasdaq.com/news/2010-07/opening-view-djia-figh... 阅读全帖 |
|
f**********g
发帖数: 2252 | 43 How long do u expect solar will keep going this run? |
|
f**********g
发帖数: 2252 | 44 How long do u expect solar will keep going this run? |
|
p****i
发帖数: 6135 | 45 【 以下文字转载自 SanFrancisco 讨论区 】
发信人: bluewater (too old, too cynical), 信区: SanFrancisco
标 题: life expectancy calculator
发信站: BBS 未名空间站 (Tue Aug 10 16:03:05 2010, 美东)
http://www.livingto100.com/ |
|
d**********o
发帖数: 1321 | 46 hw3 & hw3b 最终结果
贴一个自己 hw3b 的 match 结果让小伙伴们感受一下我们作业需要的结果是什么样子
的。
如果是同专业的小伙伴，最后拷贝到编辑器中仔细看一下，这个网页因为宽度有限，让
并排两列的比较呈现出来的很变形。
=================================================
Output of Building User Code
Explode the tar
c-.l
c-.y
makefile
scanType.h
symtab.h
symtab.c
20131116163236-(me~~)-CS445-F13-A3B.tar: POSIX tar archive (GNU)
Tests: directory
c-.l: lex description text
c-.y: lex description text
... 阅读全帖 |
|
d**********o
发帖数: 1321 | 47 hw4:老师附件
/**** Limited to 5 seconds total run time and 5000 lines of output <
<
/**** * ================================================ * <
| Tests for CS445 Assignment 4 | <
| Comparison with Expected Output | <
/**** * ================================================ * <
<
/export/home/nibbler/TestWorld <
find makefile ... 阅读全帖 |
|
d**********o
发帖数: 1321 | 48 hw3b deadline前版本
这里我也先公开承认一下自己早前犯过的错误，在这个楼的第 14 页、第 265 楼，我
有贴出自己的hw3b，但现在我必须向亲爱的读者一一您公开道歉。因为当时因为着急，
我没能找到hw3b deadline前的版本，就把这次作业后来的最终版本（也就是那个楼贴
出来的，只剩下ReturnK、CompoundK两条小尾巴的版本，那个版本被我改动了时间戳我
现在已经把它改回来了。相比于急于求成绩，我更希望自己作一个诚实的人，希望大家
能原谅）给贴出来了。因为我们作业deadline之后也还都是可以接着交的，所以任何作
业，老师都是可以根据后来的版本，追终到这个学生的最后进展的，这个最新进展老师
到底给不给分、就要看老师如何下决定了。
deadline前的版本，被我附在下面了。
当然，把自己的这个版本更新后，我愿为这个楼（这一次出来写的所有）属于自己的每
个贴子负责，所有写出来的，都是真实发生过的事。
=================================================
Output of Building User Code
Ex... 阅读全帖 |
|
d**********o
发帖数: 1321 | 49 所有测试文件
aamain2.c-
aamain2.expected
aamain2.in
aamain2.out
aamain2.tm
aamain3.c-
aamain3.expected
aamain3.out
aamain4.c-
aamain4.expected
aamain4.out
aamain4.tm
aamain5.c-
aamain5.expected
aamain5.out
aamain.c-
aamain.expected
aamain.out
aamain.tm
aasmaller.c-
aasmaller.expected
aasmaller.out
all.c-
allErrors2.c-
allErrors2.expected
allErrors2.in
allErrors2.out
allErrors.c-
allErrors.expected
allErrors.in
allErrors.out
all.expected
all.in
all.out
arglist2.c-
arglist2.expected
arglist2.out
arglist... 阅读全帖 |
|
d**********o
发帖数: 1321 | 50 最终版本的compiler测试结果
=================================================
Output of Building User Code
Explode the tar
c-.l
c-.y
scanType.h
makefile
symtab.h
symtab.cpp
emitCode.h
emitCode.cpp
20131214164956-huang-CS445-F13-A5.tar: POSIX tar archive (GNU)
Tests: directory
c-.l: lex description text
c-.y: lex description text
emitCode.cpp: ASCII C++ program text
emitCode.h: ... 阅读全帖 |
|
