由买买提看人间百态

topics

全部话题 - 话题: correlation
首页 上页 1 2 3 4 5 6 7 8 9 10 下页 末页 (共10页)
T**********y
发帖数: 157
1
来自主题: Returnee版 - 快来看牛逼的27岁教授
【 以下文字转载自 Faculty 讨论区 】
发信人: TenMilesADay (郭十迈), 信区: Faculty
标 题: 快来看牛逼的27岁教授
发信站: BBS 未名空间站 (Sun Mar 11 13:14:59 2012, 美东)
http://www.ccse.uestc.edu.cn/teacher/teacher.aspx?id=414
所有已经发表论文清单
(发表时间序)

【1】 周涛,傅忠谦,周佩玲,张建荣,张德学,”基于遗传算法的大规模流量
工程问题求解”,计算机应用,2003年第6期,43-45
【2】 杨春霞,周涛,周佩玲,刘隽,基于Multi_Agent的股市经济系统建模与
分析,自动化理论、技术与应用卷十,中国科学技术大学出版社,2003年,596-601(
中国自动化学会第18届青年学术年会会议论文集)
【3】 周佩玲,许民,赵亮,周涛,”混沌信号奇异性检测与外界冲击度量”,
数据采集与处理,Vol.19,195-198,2004
【4】 周涛,徐俊明,刘隽,”图直径和平均距离极值问题研究”,... 阅读全帖
a*****i
发帖数: 1045
2
一个Mixed model的模型。两组人类,一组有病的,一组没有病,5个时间点,主要比较
他们的mmse score 随时间的变化,covariate 有年龄和housing conditions(3个不同
的housing conditions) 这这个model里面,我们用了random intercepts and random
slopes.
1. 在开始时间段,发现两组人类,两组的mmse score的variance 一组大,一组很
小,问问 mixed model 怎么take this into count, 还有两组observations,一组明显
很多人,一组很少人。想问问mixed model count for random effects, 那么人在这里
就是random effects, mixed model 是相当于两个个组分开做linear regression的,
所以并无所谓两组人,一组的variance很大,一组的variance很小。还有两组人数不一
样的问题,unbalance data都可以解决了吗?
2. 我们做了一个cor... 阅读全帖
f**d
发帖数: 768
3
来自主题: Neuroscience版 - eBook: From computer to brain
这是一本计算神经科学的优秀著作,全文拷贝这里(图和公式缺),有兴趣的同学可以
阅读
如需要,我可以分享PDF文件(--仅供个人学习,无商业用途)
From Computer to Brain
William W. Lytton
From Computer to Brain
Foundations of Computational Neuroscience
Springer
William W. Lytton, M.D.
Associate Professor, State University of New York, Downstato, Brooklyn, NY
Visiting Associate Professor, University of Wisconsin, Madison
Visiting Associate Professor, Polytechnic University, Brooklyn, NY
Staff Neurologist., Kings County Hospital, Brooklyn, NY
In From Computer to Brain: ... 阅读全帖
r*********t
发帖数: 4911
4
来自主题: USANews版 - 和华左聊AA,聊进藤校
sure, 目前我有加州几个学校和整个加州系统的correlation, 常春藤的correlationu
florida的correlation,密歇根大学的correlation, 以及UT austin的correlation. 目
前不符合correlation的只有u georgia 和 u nebraska (其实u georgia 也是有的,
因为goergia ban了之后,黑墨反而上升,证明以前u georgia 和别人不同,他是歧视
了黑墨)。大量的correlation如果都不imply causation, 那还聊个鬼。
l******h
发帖数: 2
5
Highly reputational company with very competitive compensation for all
positions.
Please send in your resume and position/positions you like to apply to my
email address below.
h*******[email protected]
All resumes will be deliver to hiring manager directly. Act fast!
Thanks!
Silicon Engineering Group
Sr. Physical Design Timing Engineer
Timing (STA) Manager
Senior Physical Design Engineer
CAD Manager - Front-End Design and Verification
Sr. CAD Engineer - Place & Route / Physical Design Engineer
Sr. CA... 阅读全帖
l******h
发帖数: 2
6
Highly reputational company with very competitive compensation for all
positions.
Please send in your resume and position/positions you like to apply to my
email address below.
h*******[email protected]
All resumes will be deliver to hiring manager directly. Act fast!
Thanks!
Silicon Engineering Group
Sr. Physical Design Timing Engineer
Timing (STA) Manager
Senior Physical Design Engineer
CAD Manager - Front-End Design and Verification
Sr. CAD Engineer - Place & Route / Physical Design Engineer
Sr. CA... 阅读全帖
l*****a
发帖数: 38403
7
来自主题: TrustInJesus版 - 数据:基督徒在美国的数量和比例
来,这还有一些,当然这些都不够“scientific”啦
又是大神的儿子又是大神他自己的耶稣都没上过学,说不定还是个文盲(既然他老没写
过一个字),连收的徒弟多半都是干体力活的手艺人,可想而知这智商和教育程度是根
本做不得准的,对吧
The EU survey finds a positive correlation between leaving school early and
believing in a God. http://ec.europa.eu/public_opinion/archives/ebs/ebs_225_report_en.pdf
An international study has reported positive correlations between levels of
education and not believing in a deity. http://www.pitzer.edu/academics/faculty/zuckerman/Zuckerman_on_Atheism.pdf
According to a Gallup ... 阅读全帖
w*********r
发帖数: 3382
8
来自主题: TrustInJesus版 - 信基督的与被骟过的马
pssss, when did i ever CONCLUDE that there is causal relationship?
i simply pointed out that there is a correlation, and that is just plain
statistical data.
the beauty of correlation is that it indicates a predictive relationship
that can be exploited in practice, that's all what i am trying to say:
deeply religious population tends to be dumber ... that is what the data
says, :)
same applies to the butter production in Bangladesh vs S&P 500, if proved to
be a reliable correlation, you CAN use ... 阅读全帖
d******e
发帖数: 17
9
可以,直接regression, correlation就行了;
但结论仅对此人成立,且不一定和group-pooled correlation相近;
Denote Z= Individual No
因为你估计的是 corr(X1,X2 | Z=1),
Need corr(X1,X2 | Z=i)=corr(X1,X2 | Z=j), for any i, j
且 corr(X1,X2 | Z=i)!=corr(X1,X2)
所以结论是你的estimator coverge to corr(X1,X2 | Z=1) 且还需假设 n1 ->\infty.
------------------------------------------
1、即使所有人都相同的 correlation within individual, group correlation is
not equal to correlation within individuals. 这和把一个longitudinal data进行
pool算一样;
2. 需要假设individual correlation equa... 阅读全帖
C****g
发帖数: 2220
10
法律環境、公司治理與企業價值 : 基於紐交所上市電信企業的實證研究
Author: 李慧鏑
Li, Huidi
Title: 法律環境、公司治理與企業價值 : 基於紐交所上市電信企業的實證研究
Fa lu huan jing, gong si zhi li yu qi ye jia zhi : ji yu Niu jiao suo shang
shi dian xin qi ye de shi zheng yan jiu
Legal environment, corporate governance and firm value : an empirical study
of the telecommunications firms traded in NYSE
Degree: D.Mgt.
Year: 2007
Subject: Hong Kong Polytechnic University -- Dissertations.
Telecommunication -- Management.
Department: Graduate School... 阅读全帖
e****l
发帖数: 3319
11
来自主题: USANews版 - How Rich People Raise Rich Kids
转个帖子,大家看看有钱的老爹(包括收养)和自己有钱的相关性。再看看在过去三年
有钱人的财富翻倍,而低收入家庭反而减少的事实,再想想什么叫阶级固化,以及既得
利益者们如何巧取豪夺地不断扩大财富的,再想想它们扩大这些财富从哪儿来的?难到
跟穷人同期收入减少没有关联?
http://www.theatlantic.com/business/archive/2015/07/rich-people
Lately, it seems that every new study about social mobility further corrodes
the story Americans tell themselves about meritocracy; each one provides
more evidence that comfortable lives are reserved for the winners of what
sociologists call the birth lottery. But, recently, there have been
suggestions ... 阅读全帖
s***5
发帖数: 453
12
http://fivethirtyeight.com/features/election-update-why-our-model-is-more-bullish-than-others-on-trump/
As I wrote last week, Hillary Clinton is probably going to become the next
president. But there’s an awful lot of room to debate what “probably”
means.
FiveThirtyEight’s polls-only model puts Clinton’s chances at 85 percent,
while our polls-plus model has her at 83 percent. Those odds have been
pretty steady over the past week or two, although if you squint you can see
the race tightening just... 阅读全帖
S**Y
发帖数: 136
13
来自主题: JobHunting版 - 一个比较有趣的面试问题
觉得比较有趣。
鸭子喜欢在冬天晒太阳。 鸭子可能一天呆在一个草地的几个地方晒太阳。比如一开始在
坐标轴是 (1,2)的地方晒, 后来移动到 (10,10), 再后来移动到 (7,6). 研究发现,
鸭子天与天之间选的地方有correlation.比如,鸭子今天选的是
(1,2), (10,10), (7,6)
那么明天选的地方很可能离这些很近:
(1.1,2.3),(11, 10.5), (7.7,6)
发现有些鸭子correlation大些,有的小些。现在给你几天的数据,让你develop一个函
数,或者model来量化这些天与天之间的correlation(一天当中的地点不用考虑correl
ation,也就是(1,2)与(10,10),(7,6)之间的correlation不用考虑).
比如如果一只鸭子A:
day 1: (1,2), (10,10),(7,6)
day 2: (1,2), (10,10),(7,6)
day 3: (1,2), (10,10),(7,6)
day 4: (1,2), (1
q*d
发帖数: 22178
14
☆─────────────────────────────────────☆
swjtuer (swjtuer) 于 (Tue Mar 30 23:56:30 2010, 美东) 提到:
星期一
早餐 牛奶,煮鸡蛋,玉米馒头,榨菜
早点 芦柑
午餐 中大:绿豆饭,木须肉,海米冬瓜,棒骨菠菜汤
小婴:绿豆饭,三叶瓜炒肉,海米冬瓜,棒骨菠菜汤
午点 冰糖山楂水,蔬菜饼
晚餐 棒骨冬苋菜瘦肉粥,卤肉秋叶夹馍

星期二
早餐 牛奶,蒸鸡蛋,枣泥糕
早点 香蕉
午餐 中大:胡萝卜饭,芹菜炒牛肉,珊瑚豆腐,棒骨凤尾汤
小婴:胡萝卜饭,芹菜炒肉,珊瑚豆腐,棒骨凤尾汤
午点 果汁,卤猪肝
晚餐 豌豆肉末焖饭,番茄虾皮紫菜汤
星期三
早餐 南瓜粥,卤蛋,如意肉卷
早点 脐橙
午餐 中大:金银饭,咸烧白,海米儿菜,棒骨冬苋菜汤
小婴:金银饭,樱桃肉,海米儿菜,棒骨冬苋菜汤
午点 牛奶,蛋卷
晚餐 米饭,蒜苔炒肉,芙蓉蒸蛋,棒骨海带萝卜汤
星期四
早餐 牛奶,蒸鸡蛋,奶香馒头
早点 苹果
午餐 中大:南瓜饭,什锦鱼丸,韭菜炒豆干,番茄菌类鱼汤
... 阅读全帖
l****z
发帖数: 29846
15
下面就有人贴了阿.考过G的很容易理解吧.
The 'study' in question actually proved nothing... there is no corollary
here... What exactly are the murder rates using weapons purchased before and
after the 2007 repeal? In order to show causation, there must be an
accurate assessment and accurate interpretation of the statistics...
In this case, to show causality, you need to compare and contrast the
statistics using handguns purchased before (permit-required) and after (
permit-not-required) and the number of murders co... 阅读全帖
t*******r
发帖数: 22634
16
坑王的标题很有煽动性,也跟原文偏离不少。不过坑王的资料的信息量
还是不错滴。
不过这个例子倒是明显的 correlation vs causation 的问题。这个文
章里所有数据显示的都是 crime rate (或其他的 rate) 跟大片的
geography area 的 correlation。文中再把不同的 geography area
按照 belief 的情况进行 classify,来论述 crime rate 跟 belief
之间的 correlation。首先这个是很 weak 的 correlation,因为不同
的 geography area 的差别远远不止是 belief,是不是其他的因素造成
的 correlation 未可知。另外这个 causation 的问题还可以 argue 到
底是 belief 造成 geography area 的其他方面的不同,还是 geography
area 其他方面的不同造成 belief 的不同。
看着玩玩可以
h*h
发帖数: 27852
17
来自主题: Soccer版 - 胜场3分制事实上伤害了足球
http://www.cafefutebol.net/2013/09/11/why-three-points-for-a-win-is-a-loss-for-football-a-closer-look-into-one-of-the-most-important-rules-in-football-history/
Why ‘Three Points for a Win’ is a Loss for Football — A Closer Look Into
One of the Most Important Rule Changes in Football History
by NICK CHOLST on Sep 11, 2013 • 6:30 pm 7 Comments
An Introduction to Incentives
Ask an economist how to solve a problem, and he’ll tell you incentives are
the answer.1 He wouldn’t be wrong. Punishment... 阅读全帖
i******e
发帖数: 217
18
来自主题: TrustInJesus版 - 基督徒等于低智商者么
所谓个别研究无从谈起,首先这个研究本身就容易引起争议,根本没有几个人愿意去搞
。但是既然搞了,我相信他们的操守是会尽量不偏颇的,你当还是可以怀疑,只是同一
贯一样,拿不出证据而已,对不对?
其次,你也可以去搜搜别的组研究结果如何,我反正没找到同样对象样本下相反的结果
,倒是有一些相似的结果,你去搜搜
因为事实是这样,两者之间有没有correlation和怎样的correlation,对同样的对象群
是一个既定事实,所以只要方法足够科学客观,样本足够大,结果肯定是一样的
wiki里后面的arguement,同写paper一样,总要对自己的方式方法进行总结和critical
thinking。有些不足的地方指出来,没错。但是paper本身的基调已经定了,结果就在
那里,你去argue好了,反驳好了,但是要拿出数据来。你有么?没有的话就不要妄谈。
而且关键的问题根本不在两者之间的correlation,对数据的直接翻译就是两者之间有
strong correlation,没有什么好argue的。你要argue的话应该去说明虽然有
correlation,但是没有causality。涉及到人和... 阅读全帖
S**Y
发帖数: 136
19
来自主题: Programming版 - 一个比较有趣的面试问题
觉得比较有趣。
鸭子喜欢在冬天晒太阳。 鸭子可能一天呆在一个草地的几个地方晒太阳。比如一开始在
坐标轴是 (1,2)的地方晒, 后来移动到 (10,10), 再后来移动到 (7,6). 研究发现,
鸭子天与天之间选的地方有correlation.比如,鸭子今天选的是
(1,2), (10,10), (7,6)
那么明天选的地方很可能离这些很近:
(1.1,2.3),(11, 10.5), (7.7,6)
发现有些鸭子correlation大些,有的小些。现在给你几天的数据,让你develop一个函
数,或者model来量化这些天与天之间的correlation(一天当中的地点不用考虑correl
ation,也就是(1,2)与(10,10),(7,6)之间的correlation不用考虑).
比如如果一只鸭子A:
day 1: (1,2), (10,10),(7,6)
day 2: (1,2), (10,10),(7,6)
day 3: (1,2), (10,10),(7,6)
day 4: (1,2), (1
h***y
发帖数: 834
20
☆─────────────────────────────────────☆
LiuDier (Dier) 于 (Thu Jun 23 20:14:50 2011, 美东) 提到:

点好玩的八卦?呵呵
遵版主吩咐:
☆─────────────────────────────────────☆
LiuDier (Dier) 于 (Thu Jun 23 20:19:05 2011, 美东) 提到:
Case 1. 发生在某ACGME accredited residency program的对话, 原文是英文:
Dr. Nightlife:C国首都冰城, 曾爆发一起食物A中毒事件,导致死伤多人,C国政府
的独裁统治应该负全部责任!
Dr. Carlos:(随手查了查PubMed)在美国也有非常类似的疫情发生啊?而且不止一起
,就譬如《NEJM》报道的内华达州Tenk中毒事件,以及华盛顿DC的XXX事件…
…美国可是民主政体啊。所以,食物A中毒致死事件,不能说是C国特色,也不能把发
生机制完全归咎于C国政府的独裁统治,还应该从两国政府的相同点上寻找原因,譬如
说食品... 阅读全帖
w**********y
发帖数: 1691
21
多谢分享.大概做了做..欢迎补充和指正.
- sqrt(i)=?
e^{\pi/4 i} or - e^{\pi/4 i}
- You and me roll a dice,first one gets a six wins. You roll first. what
is the probability of you winning?
P(I win) = P(Y !win and I win) = 6/11
- A stair of n steps. Each time you step up 1 or 2 steps. How many
different ways are there to reach the top? what is the asymptotic limit?
Fibonacci sequence ..limF(n)/F(n-1)==x for n>2, solve x, and F(n) ~ x^{n-1}
- Moment generating function of standard model.
statistic book…
- Write a si... 阅读全帖
t*******y
发帖数: 637
22
第二题应该是6/11吧
能讲讲这个吗? - X1 and X2 are independent random variable with pdf f and g.
what is what is the pdf of X=X1+X2
Jacobian matrix for X1+X2 and X1-X2..

多谢分享.大概做了做..欢迎补充和指正.
- sqrt(i)=?
e^{\pi/4 i} or - e^{\pi/4 i}
- You and me roll a dice,first one gets a six wins. You roll first. what
is the probability of you winning?
P(I win) = P(Y !win and I win) = 5/6*1/6
- A stair of n steps. Each time you step up 1 or 2 steps. How many
different ways are there to reach the top? what is the asymptotic... 阅读全帖
D***R
发帖数: 10
23
芝加哥一家prop shop,面试官是一个trader,先问了我在学校做的一个项目,很详细
,然后就是这两个独立的题目
1. 有两个random variables x1, x2, correlation(x1,x2)=0. 现在 linearly
regress y on x1,x2, 归化结果是 y_hat. 又知道 correlation(x1,y)=0.5,
correlation(x2,y)=0.6. 问correlation(y, y_hat)=?
2. 有两个random variables x1, x2, correlation(x1,x2)=0.98. 现在 linearly
regress y on x1,x2, 怎么做,有什么问题?我回答colinearity, 又问如何处理,我
说了两个办法,一是PCA,二是regularization方法比如Ridge和lasso。又问ridge和
lasso在这里用哪个好,为什么?我就卡住了,仅上过一门machine learning课,还是
基本知识太差啊。
S**C
发帖数: 2964
24
Well, REIT is high correlated with Mid-cap.
05/22/1998 - 04/30/2015 based on daily returns, VGSIX and VIMSX have 71%
correlation, and their return is about the same, 9% annualized.
Since 01/01/2002, their correlation is 77%. VGSIX perform better by 0.35%
annualized than VIMSX, with 40% higher SD, not an attractive trade-off. From
diversification point of view, REIT hardly provide much protection because
its correlation with midcap.
The good thing I can say about housing is it is local and very ... 阅读全帖
D******e
发帖数: 36
25
来自主题: Parenting版 - 我来说一下天才的鉴定方法
女人只要低B格,都是会popular的,与是不是天才没有关系。
广州火车站,5块盒饭钱就可以玩一次,很多民工去玩,那些女的也很popular啊。
popular与天才没有correlation,甚至是反correlation。至少对我是负correlation,
因为跟天才女生的孩子,与父母的iq没啥关系,甚至父母越傻,生的孩子说不定是天才
,所以老话说聪明劲别用光,留下些给子孙。
popular与颜值和B格有correlation。
你认为自己popular,可能与自己的低B格或者高颜值有关。我不知道是哪个。
G****n
发帖数: 145
26
长江各领域教授论文引用次数
Between trust and control: Developing confidence in partner cooperation in
alliances
[PDF] from cuny.edu…, BS Teng - Academy of management Review, 1998 - JSTOR
Strategic alliances have been recognized as arenas with potential for
opportunistic behavior
by partners. Hence, a firm needs to have an adequate level of confidence in
its partner's
cooperative behavior. In this article we examine the notion of confidence in
partner ...
Cited by 1749 - Related articles - BL Direct - All 9 v... 阅读全帖
Y******u
发帖数: 1090
27
来自主题: Stock版 - nikkei和spy的相关关系
Nikkei is better correlated with the inverse of Yen, which is correlated
with Dollar, which is again inversely correlated with spy.
The inverse correlation between dollar and spy didn't hold in July, when
both rose. That's where the divergence starts to show up in your chart.
s***m
发帖数: 6197
28
来自主题: Stock版 - 答谢文,谈谈VIX 和 VIX future
【 以下文字转载自 Quant 讨论区 】
发信人: throwaway (专门注册), 信区: Quant
标 题: 答谢文,谈谈VIX 和 VIX future
关键字: VIX
发信站: BBS 未名空间站 (Sun May 11 18:02:53 2014, 美东)
上次发文求建议,收到很多真诚的帮助。想了想可以写篇我对 VIX index 的学习心得
,算是表达感谢吧。
说明一下,我毕竟是个菜鸟新人,这篇也只是综合公开资料再加上我自己对 VIX 的理
解,也就希望对准备学习这个领域的人有所帮助,熟悉volatility trading的内行看了
肯定觉得简单,有不对的地方请多多指正。
废话完毕。
1. Rationale behind VIX
想了解 VIX index 计算方法的依据,必读的文章是
“More than you wanted to know about variance swap".
http://elis.sigmath.es.osaka-u.ac.jp/research/gs-volatility_swa
具体的证明比较繁琐,我总结就是:
In a n... 阅读全帖
T**7
发帖数: 264
29
请发给我:
你的
(1)简短的PUBLICATION LIST
不超过5篇文章
(2)你的名字
(3)学位
(4)EMAIL地址
谢谢
entitled "Serum soluble CD30 - a marker of airways inflammation in multiple-
trigger wheezers?"
Article Type: Original Article
Background: Imbalance in activation of T-helper cell types 1 (Th1) and 2 (
Th2) may have a link with the development of asthma. Objective: To evaluate
the role of serum soluble markers of Th1- and Th2-lymphocyte activation with
regard to lung function, inflammatory markers, and therapy response in 90
pre... 阅读全帖
n****e
发帖数: 4990
30
I didn't say LBJ IS NOT negativelyu correlated.
The origin of the post is because I said in an earlier post Rose's FTA and
Win percentage isn't negtively correlated ,like KB's data. LZ gave the data, and it
shows the difference.
Rose 's data dosen't show the negative correlation, would you agree?
KB' data show the negtive correlation , would you agree?

essentially
C*******I
发帖数: 339
31
我觉得30楼的这个例子说的是correlation 和 Causality的关系,有correlation不表
明一定有causality。这个观点我是绝对同意的,其实也正是对correlation和
causality的混淆,产生了很多以吸引眼球为目的的伪科学,鄙视一下先。
对饭后站对减肥有没有效果这个问题,首先有没有correlation都还没法确定,就更不
要谈causality了。
但是我其实还是不大明白这和null hypothesis 有什么联系。
b*s
发帖数: 82482
32
来自主题: LeisureTime版 - 副校长的枪
建议你读一下Nassim Nicholas Taleb的《The Black Swan: The Impact of the
Highly Improbable》。其中一点就是,只要你留意去找,世界上所有得事情都可以找
到Correlaltions的,所有的事情都可以解释。所以,事后诸葛亮特别容易,一战世界
领袖都没有看出来战争的必然,二战盟国和苏联都是大笨蛋,“明显”的correlation
都看不出来。
事后这些个学者们就特别牛掰的指点江上了……
经济危机以后的大牛们也冒出来指点correlations,唉。都是聪明人啊
这个《Black Swan》看过以后,你就会重新审视这些个社会科学和他们的方法论。你就
会为什么很多人都认为社会科学都是蒙事的了……

我从来没有说必然联系,社会上的事哪来那么容易的必然联系。我强调的是
correlation,这个对现在做政治决策一般都是要看的。连 correlation 和 causality
都分不清楚而且只是放个例的你叫我说啥呢?
题。
-
m********n
发帖数: 3558
33
你常看英文的影评什么的吗。我外行看看感觉写得很专业。可是为什么要用英文写呢?
sensory-motor correlation 何解啊?只知道sensory-motor system. (correlation
不是都是A 和 B 之间的correlation me , 这个是说 sensor 和 motor 之间的
correlation,还是什么别的?)
是为了不想剧透,所以没说具体情节吗?
b*s
发帖数: 82482
34
Li's paper is called "On Default Correlation: A Copula Function Approach" (
2000), published in Journal of Fixed Income, Vol. 9, Issue 4, pages 43–54.
In section 1 through 5.3, Li describes actuarial math that sets the stage
for his theory. The mathematics are from established statistical theory,
actuarial models, and probability theory.
In section 5.4, he uses Gaussian Copula to measure event relationships, or
mathematically, correlations, between random economic events, expressed as:
In layman... 阅读全帖
l*****o
发帖数: 26631
35
Long-Term Health Risks and Benefits Associated with Spay / Neuter in Dogs
Laura J. Sanborn, M.S.
May 14, 2007
Precis
At some point, most of us with an interest in dogs will have to consider
whether or not to spay / neuter our pet. Tradition holds that the benefits
of doing so at an early age outweigh the risks. Often, tradition holds sway
in the decision-making process even after countervailing evidence has
accumulated.
Ms Sanborn has reviewed the veterinary medical literature in an exhaustive
a... 阅读全帖
l******8
发帖数: 9475
36
如果研究结果是可信的, 你还想把自家的狗去势吗?
http://www.caninesports.com/SNBehaviorBoneDataSnapShot.pdf
======================================================================
重点摘抄:
By using large a sample of dogs than any used previously to examine behavior
in dogs, we found significant correlations between neutering dogs and
increases in
aggression, fear and anxiety, and excitability, regardless of the age at
which the dog was neutered. There were also significant correlations between
neutering and decreases in tr... 阅读全帖
p***l
发帖数: 134
37
来自主题: TrustInJesus版 - 再来一个数据——信教和IQ的关系
统计数据分析及其结果的解释真的不是非专业人士想象的那样容易的。大家都知道
correlation不等于causality,于是试图创造出各种看似荒谬的演示图表。比如这
个“柠檬vs交通事故死亡率”,其实在统计学家看起来一点儿都不可笑。当你观察
到两个变量高度相关(r^2很大),那么它们之间一定是有某种合理联系的。我们要
做的不是指着这个图表大笑:“看!纯粹的统计计算多么的荒谬!”而是要仔细思
考一下,这个被观察数据揭示出来的自然、社会现象乍看起来为什么如此地荒谬?通
常思考的结果都将指向某种简单而且合乎逻辑的解释,而非出于偶然的契合。在我
的统计分析经验里面,还真没有见过纯粹出于偶然而且无法解释的高度相关性。
那么交通事故死亡率为什么与柠檬进口量高度相关呢?答案不再图表的横轴上,也不
在纵轴上,而是在图表里面——缺失的一环是“时间”。大家留心看一下,数据点
的时间标示几乎是线性递增的。所以对图表的解读非常简单。从1996年到2000年之
间,美国从墨西哥进口的柠檬数量按年线性递增。为什么是“线性”递增?大概因
为柠檬进口都是有计划的,而人为的计划为了方便起见都采用线性模型,所以最后
的... 阅读全帖
S**U
发帖数: 7025
38
接近。
我认为用统计学 auto-correlation 可以解释为什么会把一个变化的五蕴想成是个体
http://en.wikipedia.org/wiki/Autocorrelation
auto-correlation 值接近1时, 我们可以清楚辨别个体虽变,还是同一个(我或ta)。
auto-correlation 值接近-1时, 我们就认为是不同的个体,或是(如同学会时发现)
「ta 不像从前的ta了」。
P.S. 这是 N variable auto-correlation, N >> 1

好比: 没有你爸爸妈妈,怎么有你? 这是否就是所谓的缘分呢?
s*****0
发帖数: 357
39
临周末前说说correlation和regression吧. 很多教科书上都把这两个techniques放在
一起讨论,初学统计的人难免会觉得两者密不可分。其实两者的区别还是很明显的,服
从于不同的分析目的,很少有人对数据同时使用这两种方法。
先说说correlation。Correlation是用来检测两组continous的数据间有没有
association(另一种方法前文已提到,回顾一下)。比如年龄和脂肪含量,胳膊长和
腿长,它们之间有相关性,但没有明确的causality关系。难道说年龄是导致脂肪含量
增高的根本原因?我看缺乏锻炼吃得太多才更是原因。
测量相关性用的一个系数叫做correlation coefficient。标准的计算方法经常归功于
Pearson,所以很常见Pearson's r的提法。其范围介于-1到1间。正值为正相关,负值
为负相关,而接近于0的话则相关性不显著。通常将两组数据用scatter plot画出来能
看出端倪, 数据间的分布可能有一个underlying linear trend,而Pearson's r就是
一个测量scattered的数据
D**g
发帖数: 739
40
补充一下correlation的误用。生物试验或医学研究里经常会有这种情况:同一SUBJECT
(人,老鼠或者well of 细胞)在N条件下重复测定同一种东西。比如膜片钳做例子通道
,在同一膜片上梯度增加某种离子浓度,然后看通道电流的变化,如此做N个膜片,然
后想看电流跟离子浓度是否相关。如果用一般的回归或ANOVA方法求r是错误的。这里需
要区分WITHIN SUBJECT effect 和 between subject effect。 两者回答的问题是不同
的。有兴趣的请参见:
BMJ 1995;310:446 (18 February)
Statistics notes
Calculating correlation coefficients with repeated observations: Part 1--
correlation within subjects
BMJ 1995;310:633 (11 March)
Statistics notes
Calculating correlation coefficients with repeated observat
w****y
发帖数: 2501
41
来自主题: ChineseMed版 - 中医是蒙昧时代的产物
中医讲 coincidence
西医讲 causality
A发生了,B随之发生,这是 coincidence
A发生了,B往往发生,这是 correlation
A导致了B发生,这才是 causality
coincidence多,correlation少,causality更少。coincidence里只有一
小部分是causality。
揉脾,然后头皮屑消失可能只是 coincidence,不一定是 correlation,更不
一定是 causality。
西医通过coincidence发现方法(当然通过其他的途径),通过临床试验研究
correlation,最终目的是证明有没有causality。
中医基本上就是直接拿coincidence当causality,结果自然对少错多。这个其
实也是人类思维的弱点,人容易犯的一种错误。
G****n
发帖数: 145
42
来自主题: Economics版 - 长江各领域教授论文引用率
Between trust and control: Developing confidence in partner cooperation in
alliances
[PDF] from cuny.edu…, BS Teng - Academy of management Review, 1998 - JSTOR
Strategic alliances have been recognized as arenas with potential for
opportunistic behavior
by partners. Hence, a firm needs to have an adequate level of confidence in
its partner's
cooperative behavior. In this article we examine the notion of confidence in
partner ...
Cited by 1749 - Related articles - BL Direct - All 9 versions
A reso... 阅读全帖
w**********y
发帖数: 1691
43
来自主题: Quant版 - MS新题一题
记得去年有人讨论过.一个很nb的直观解法.
把每个random variable对应成空间中的一个向量,两个rv的correlation就是向量间夹
角的cos(余弦).
那你可以想象着,如果保持两两向量的correlation相等,就可以把它们想象成撑起一把
雨伞的三条主轴.你把伞收起来的时候,对于correlation==1.逐渐撑开,correlation逐渐减小(夹
角变大),直到三条轴在一个平面上,对应r=cos(2pi/3)=-1/2.
所以r可以取-1/2到1的任何值..
e*******6
发帖数: 13
44
来自主题: Quant版 - MS新题一题
这个解法太帅了!
=======================================================================
发信人: weekendsunny (醉生梦死), 信区: Quant
标 题: Re: MS新题一题
发信站: BBS 未名空间站 (Wed Feb 24 12:11:28 2010, 美东)
记得去年有人讨论过.一个很nb的直观解法.
把每个random variable对应成空间中的一个向量,两个rv的correlation就是向量间夹
角的cos(余弦).
那你可以想象着,如果保持两两向量的correlation相等,就可以把它们想象成撑起一把
雨伞的三条主轴.你把伞收起来的时候,对于correlation==1.逐渐撑开,correlation逐
渐减小(夹
角变大),直到三条轴在一个平面上,对应r=cos(2pi/3)=-1/2.
所以r可以取-1/2到1的任何值..
i**w
发帖数: 71
45
背景:fresh physics PhD
还没offer,但年前基本上不会再折腾了。
有重复。基本上都是很标准的题。
简单的题如果人家想问倒你也是很容易的。
面试书recruiter推荐
1) Mark Joshi: "Quant Job Interviews: Questions and Answers". I have
heard very good things about this book.
2) Xinfeng Zhou, "A Practical Guide to Quantitative Finance Interviews"
个人觉得非常有用, 大部分问题都在这两本上。
算法,C++, stochastic calculus 就看比较标准的几本。
- sqrt(i)=?
- You and me roll a dice,first one gets a six wins. You roll first. what
is the probability of you winning?
- A stair of n steps. Each time you st... 阅读全帖
l******i
发帖数: 1404
46
来自主题: Quant版 - 【Probability Problem】面试题
Transformation Method and Accept-Reject Method
are two basic ways to generate random variables.
To summarize:
1. To generate iid uniform firstly:
Most programming languages have the ability to generate pseudo-random
numbers which are effectively distributed according to the standard
uniform distribution.
For instance, In C++, use srand().
2.
Given \rho to generate two N(0,1) with correlation \rho:
From 1, we can easily generate two iid U(0,1)
random variables P and Q;
Use Box-Muller transformati... 阅读全帖
n**r
发帖数: 1587
47
来自主题: Quant版 - job opportunities
post for my recruiter, please contact them directly.
f******************[email protected]
Contact : 201 468 0519
Statistical Modeling-(PhD) - New York
Our clients, top-tier Management consulting firms advising international
banks and other financial firms based in NY, are looking for PhDs with
strong statistical modeling skills to join Quantitative Research & Analytics
teams.
The group conducts extensive research and builds prototype models that
estimate the economic stress on the credit risk of struc... 阅读全帖
t*******y
发帖数: 18
48
来自主题: Quant版 - 答谢文,谈谈VIX 和 VIX future
上次发文求建议,收到很多真诚的帮助。想了想可以写篇我对 VIX index 的学习心得
,算是表达感谢吧。
说明一下,我毕竟是个菜鸟新人,这篇也只是综合公开资料再加上我自己对 VIX 的理
解,也就希望对准备学习这个领域的人有所帮助,熟悉volatility trading的内行看了
肯定觉得简单,有不对的地方请多多指正。
废话完毕。
1. Rationale behind VIX
想了解 VIX index 计算方法的依据,必读的文章是
“More than you wanted to know about variance swap".
http://elis.sigmath.es.osaka-u.ac.jp/research/gs-volatility_swa
具体的证明比较繁琐,我总结就是:
In a nut shell, if we can ignore friction, an out-of-money option portfolio
with certain weights could replicate a variance swap, whose payoff = ... 阅读全帖
s**c
发帖数: 1247
49
来自主题: Statistics版 - 一些初级time series问题请教
q看correlation
p看partial correlation
如果是AR(p),那么partial cut off at lag p,correlation是tail off
MA(q),那么correlation cut off at lag q, partial tails off
ARMA(p,q)的话都是tail off
这个都是指理论值,实际求出来的estimate会有差异的
首页 上页 1 2 3 4 5 6 7 8 9 10 下页 末页 (共10页)