s****y
发帖数: 109 | 1 How to capture free boundary of a American Option by the least square
simulation? My solution is to solve the equation "Exercising Value =
Continuing Value" where the Exercising Value EV(S)=S-K; Continuing Value CV(
S)=a+b*S+c*S^2, where the coefficient a, b, c is estimated by least square
regression (suppose the base function is polynomial). Is this correct?
Thanks a lot. |
|
J******d
发帖数: 506 | 2 How about simulate W(t)? where W(t) is Wiener process. |
|
s******a
发帖数: 184 | 3 在 matlab 里如何simulate 符合下面这样的hierarchical distribution 的data:
X|P ~ binomial(P)
P ~ beta(\alpha, \beta) |
|
w****a
发帖数: 155 | 4 large scale Monte Carlo simulation都会遇到哪些技术难点？ |
|
k********o
发帖数: 4 | 5 哪位大侠能说说准备花街的面试，simulation 和algorithm要看什么东西？
哪本书？哪几章？
谢谢了。 |
|
s*******0
发帖数: 3461 | 6 simulation: Ross 的统计模拟 以及 Paul Glassman 的 monte carlo method in
finance
numerical analysis： 随便条一门 都差不多 呵呵
personal suggestions |
|
i**M
发帖数: 108 | 7 Suppose I have 500 curves coming from monte carlo simulation of stock price.
How to analyze these curves? Please give me some general idea. Thanks a lot
!!!! |
|
w*******1
发帖数: 112 | 8 random number generator generates standard normal variable x. how to use
monte carlo simulation and importance sampling to find the prob(x>10)? thank
you very much |
|
T*****e
发帖数: 40 | 9 X is N(0,1) random variable, 如何用Montcalo Simulation 计算概率P(x>10)?
Thanks! |
|
w**********y
发帖数: 1691 | 10 你自己做个simulation试试，2分钟的事情。你甩几万个数据看看？或者你甩几亿的数
据看看？
然后你fail了之后试试求 x>5, 然后甩几百万的数据，甩一百次看看你求出来的p的
variance。再甩一百次几亿个看看你的variance？ |
|
l**********n
发帖数: 72 | 11 Actually Matlab is not good in terms of simulation but it generates figures
quickly and provides you easy matrix manipulations. You don't even need to
worry about the diagonalization of the matrices. It really depends on what
kind of research you are in, if you don't care many details and just want to
get the result quickly, that will be just fine. |
|
s*r
发帖数: 2757 | 12 simulation is like cooking
there are only some principles, all the rest depend on personality
try tasting the flavor of 'monte carlo', it has nothing to do with routine |
|
j*****e
发帖数: 182 | 13 SAMPLE SIZE ESTIMATION AND POWER ANALYSIS. WHEN EVER YOU DO BOOTSTRAPPING,
YOU NEED MONTE CARLO SIMULATION. |
|
m**u
发帖数: 2 | 14 最近要写篇论文，导师建议念念simulation的东西，感兴趣的话也可以试着做
bootstrap和
jackknife的相关研究，不必很难。
请问大家，有没有什么好的资料可以推荐，论文什么的。
我是新手，不知道从何开始。3个半月的时间，想系统学习一下。
谢谢！！ |
|
p*s
发帖数: 71 | 15 who knows how to simulate a distribution with given mean, standard deviation
skewness, min and max. I tried to use Gamma distribution but the generated
numbers cannot reach min and max. Someone suggested me to use the mixture of
uniform distributions, but it is very difficult to find the parameters of
them. I wonder who know a simple way to do so.
Thanks! |
|
t**i
发帖数: 688 | 16 With given min and max, you can not really simulate a bona fide Gamma sample
. |
|
S******y
发帖数: 1123 | 17 I have a simulated market with 5 firms, starting with 20% market share each.
I would like to have one page of ppt to show the dynamic increase/decrease
of these five firms' market share, base on certain very simple assumptions.
(e.g., half of the customers left company A, and joined company B..)
Does anybody know how to do an "animated" page of Powerpoint to show market
share increase/decrease between five firms?
Thanks. |
|
h*******d
发帖数: 272 | 18 真诚求书 中文版的 ROSS
中文翻译一般是 《统计模拟》 罗斯
纸版的 电子版的都行
谢谢大家 |
|
h*********o
发帖数: 151 | 19 有大大遇到过这样的问题么？
用LOCAL LINEAR方法作simulation，总是出现当N增大时，estimate的MADE也增大。
查了程序，应该没有错啊！不知道有可能是哪里出了问题？ |
|
s******a
发帖数: 388 | 20 我要用SAS 计算IMPLIED VOLATILITY， 用到SIMULATION， 哪位XDJM 做过这方面的东
西请指点一下 |
|
j******o
发帖数: 127 | 21 没有这方面的经验。你是说你的model中day, plate(day), column and row都是随机变
量？你知道所有变量的distribution? 你也知道所有的parameters?然后你想simulate
the sum？对吗？ |
|
j******o
发帖数: 127 | 22 So the residual is only random variable in your model. With the data you
have, you can get residuals using your model. Then the mean and std of
residuals can be calculated. thus, sampling from normal distribution with
parameter you estimated could be the place that your simulation starts from.
as
data
the |
|
d*******1
发帖数: 854 | 23 thank you. you are right that, in this case, residual is the only random
variance. if we know the sample mean is u and we can estimate fixed effect
for day1 is d1 and fixed effect for plate 1 is p1 and residual variance is
sigma, then we can just simulate the experimental data collected on day1 and
plate 1 by generating random number from N(u+d1+p1, sigma).
however, I now realize that it would be more appropriate to treat day and
plate as RANDOM effect, so I run mixed model on sample data and ob |
|
W**********E
发帖数: 242 | 24 如何SIMULATE SURVIVAL TIME 并能按预定比例CENSOR（２５％，５０％，７５％）？
比方说TIME TO DEATH ～EXPONENTIAL DIST（RATE=XXX），然后CENSORING TIME～
UNIFORM（０，C）。如何确定这个C最后的数据有预定的CENSORING PROPORTION（２５
％，５０％，７５％）？
谁有这方面的经验
THANKS， |
|
l***o
发帖数: 194 | 25 I want to simulate a data matrix using the 3PL and Generalized partial
credit model.
I am wondering whether anyone knows the distribution of the a, b and c
parameter of the 3PL and their range.
also the distribution of a and b and range of the generalize partial credit
model.
if anyone can give me some help on this, I will really appraciated.
mimi |
|
W**********E
发帖数: 242 | 26 No, I need covariates X to simulate outcome Y because Y are also associated
with Xs plus treatment group. The settings are common for propensity score
modeling.
There are no problem with model Y=Xs Group. The estimates for X and Groups
are very close to true parameters.
There are problems with model Y=group propensity. So I think it might be
something wrong with the way I generate group and propensity.
奇怪的就是很多文章都是说PROPENSITY SCORE ADJUSTED模型会好，估计的系数会有更小偏差 |
|
W**********E
发帖数: 242 | 27 问题不是模型复杂与否，我只是给了个例子。问题是调整PROPENSITY SCORE后，竟然
TREATMENT的系数估计会有更大的偏差，这和其他文献的发现相反，所以想知道哪里出
了偏差。可是我看不少文献就是一样的方法做SIMULATION
x10 |
|
b**********i
发帖数: 1059 | 28 #simulation =200 太小，先调到 2000看看 |
|
b**********i
发帖数: 1059 | 29 照我下面的做，你给个包子吧。
在loop外面加个loop，1000/500都随你。然后求个group effect 的mean。 monticarlo
simulation parameter estimate 也是random的不是。 您老就搞一个dataset就想歪
诗大街啊。 |
|
W**********E
发帖数: 242 | 30 我只是给个SAS例子。原来的SIMULATION我在R里做了２０００次，不过是算TYPE I
ERROR RATE 和POWER。TREATMENT GROUP的P－VALUE ON AVERAGE两个模型（adjusting
ALL covariates vs adjusting for PROPENSITY SCORE）差很多,adjusting for
propensity score的POWER低很多. 我才仔细看单个数据的ESTIMATE，发现很多次同一
个数据两个模型TREATMENT GROUP的结果很不一样，这让我觉得很费解。实际分析数据
中我觉得调整ALL X或PROPENSITY SCORE差别是有但不会太大。就算PROPENSITY SCORE
的ESTIMATE结果 on average 接近TRUE PARAMETER，但POWER比ADJUSTING ALL X还低很多，还是很难
理解。
如果没人觉得我产生GROUPING的方法有问题，那到好了。包子给你，多谢帮忙
monticarlo |
|
a********6
发帖数: 14468 | 31 啥分布产生standard deviations用于simulation 较靠谱？因为SD都要求大于0,
lognormal的更要求大于1，正琢磨呢？谢谢。 |
|
f***a
发帖数: 329 | 32 monte carlo simulation里的probability 数据
你到底指啥？ |
|
a****3
发帖数: 11741 | 33 我想问问Monte Carlo Simulation和probability有什么关系？ |
|
a****3
发帖数: 11741 | 34 我在看讲义，提到“estimate the probabilities", 接下来又说：
Where does the data come from?
1. Primary data from studies
2. Literature Reviews
3. Expert Opinion
4. Consensus development conferences
5. Best Guess
6. Local vs Global values
然后讲到Monte Carlo Sensitivity Analysis:
Probabilities that calibrate standard decision trees are point estimates of
a quantity: they do not contain
estimates of the accuracy or precision of that variable.
我想我大概明白我要怎么用monte carlo simulation了。 |
|
b**********i
发帖数: 1059 | 35 谢谢，还得学习一下macro，不知道怎么写loop。能不能请你给几行code让我看看大概
是咋样。我悟性还不错的。
log file 存到盘上了。等关了sas才能看。
刚刚看了一下specification，晕我把simulation 次数设成10000. 这个实在是。。。。
什么时候sas能用上多核啊？
just |
|
w****a
发帖数: 155 | 36 large scale Monte Carlo simulation都会遇到哪些技术难点？ |
|
d**********r
发帖数: 24123 | 37 有阿，一些复杂的Design我就直接用simulation 去估计sample size了。 |
|
a********a
发帖数: 346 | 38 Then how to simulate if it is non normal data? Thanks |
|
H**********J
发帖数: 351 | 39 帮上海罗氏推荐了一圈人，方向都不对，麻烦大家帮忙看看，我也好重新确定下方向，
谢谢大牛们了！！
工作内容：
Responsibilities:
•will be responsible for statistical analysis in projects from target
identification to proof of concept, with an emphasis on translational
medicine projects using high dimensional molecular data,
•will work as a member of interdisciplinary project teams, but also
provide consultancy and service on request for specific projects. Typical
tasks will include the design of experiments, power analysis and development
... 阅读全帖 |
|
H**********J
发帖数: 351 | 40 没industry经验的30万左右，有industry经验的看人定；
是统计还是Phase 1 PK & PD Simulation & Molding ？ |
|
H**********J
发帖数: 351 | 41 信息更新！
刚刚跟罗氏line Manager沟通完回来，要点如下：
1，不招新手，一定要industry经验的，越资深越好；薪酬可谈；
2，方向有2个：
A：Biostatistician of early stage Clinical Trail (Phase 1 and Phase 2A);-
location 上海；招1人；
B: Biostatistician background with experience in Phase 1 simulation and
molding;-Location 北京；招1人；
有大牛有兴趣的，联系发简历我：j****[email protected]
谢谢。 |
|
l****a
发帖数: 352 | 42 就是phase1, phase2a PoC, and biomarker/genetic marker exploratory research
三合一的statistician，做点pkpd modeling and simulation，做点bayesian
adpative design for clinical trial
三样全占的不多见，全都做得精的更加少，得看罗氏开多少钱了:)
小打击一下，fresh phd能符合要求的基本没有 |
|
s*****0
发帖数: 357 | 43 工资开低了，符合原文条件的在北美交的税都不只这些。这年头又懂clinical trial又
明晰genetics/pharmacogenomics熟悉diagnostics的人如果没有身份问题就是香饽饽，
recruiter的电话不会断。
我猜写job description的人糊涂了，把各种需要堆砌在了一起，应该精简一下，比如
做Phase 1/2的具体要求什么，prefer什么，而做simulation的又要求什么知识通晓什
么平台构建过什么模型。如果仅仅要求前期临床的experience而不加入bioinfo/comp
bio的要求，这价钱倒还适中。有些知识可以后期培养，公司要求一步到位就别吝惜花
钱。 |
|
g******e
发帖数: 140 | 44 Hello,
I have one series of observations on a ratio variable. after I plot the
histogram, it does not look like those common PDFs. Is there a way that I
can run monte carlo simulation just based on these observations. I was
thinking to run proc kde to a get density estimation, but don't know how to
generate the RV based on the output of proc kde.
Thanks, |
|
x******r
发帖数: 665 | 45
~~~~~~~~~~~~~~~ :monte carlo就是你simulate数据时用
的一种方法而已，这两者是包含关系吧 |
|
s*********y
发帖数: 34 | 46 最近需要用Monte-Carlo simulation in finance, 不时很明白这个方法. 有哪位大侠
给介绍一下.
我的目的是要算出一个产品针对某种客户群的代表价格. 现在我有的以前做的资料是,
抽取5000客户的信息, 计算客户针对某一种产品的5000个价格, 就会有一个价格的分布
,以前是假定正太分布,用样本的mean /median价格做代表价格. 如果这个分布可能不
是正太. 怎么能决定针对种客户群代表价格. |
|
l******n
发帖数: 9344 | 47 现在不假定normal的话，那就是某种其它的distribution,比如exp,beta之类的，然后
你再fit
没看出来哪里要simulation
, |
|
|
K****a
发帖数: 67 | 49 请版上好心人帮帮我一下，感激不尽
需要用SAS Monte Carlo simulation，模拟一组实验数据 group randomization trial：
要求outcomes within a group are correlated and binary(0/1)，given
correlation coefficient as rho=0.01.
predictor variables 有categorical (e.g. treatment, gender ....)和continous
类型。
我搜了许多文献，许多写道要用beta-binomial distribution 来生成数据，可是依然
没有弄明白怎么写code,把这个outcome的相关系数rho插进去，或者我怎么确定相关度
是我想要的呢？
谢谢各位
这是网上一个例子：
data test;
>> b0 = -1.3;
>> b1 = 2.2;
>> b2 = 0;
>> s = 0.6;
>> do group=1 to 30;
>> gamma = s*rannor(123... 阅读全帖 |
|
h*********n
发帖数: 278 | 50 Say I have a model predicting Y which is a dollar amount. And I want to use
Monte-Carlo simulation to obtain a confidence interval of Y. Does anyone
have any suggestion on how to do it or where to learn?
Many Thanks! (sorry can't input Chinese at work) |
|
