P****d
发帖数: 113 | 1 Stochastic Process might help you understand sequential analysis.
slightly more useful in biostat than time series analysis. |
|
y***t
发帖数: 644 | 2 Also, Meta-analysis methods by Hunter & Schmidt.
Books by Morgan Wang and Bushman or Arthur, Bennett, Huffcut have SAS code
for meta-analysis. |
|
l***a
发帖数: 12410 | 3 then factor analysis? no need to do multicolinearity analysis? |
|
F****3
发帖数: 1504 | 4 请问哪位高人有Understanding Analysis （作者Stephen Abbott ，Springer-Verlag 出
版)的Solution Manual 吗？今年秋天要读Econ 的MWG那本书。里面有很多Analysis，
根本看不
懂，痛哭流涕。
想看看分析补补课。万分感激！
我的邮箱是 kazaaerato [at] gmail.com |
|
s******a
发帖数: 184 | 5 对于连续变量做Bayesian analysis, likelihood function往往是0或者非常小的数.程
序往往就blowup 了或根本产生不了什么有意义的结果。
请问有没有数值上的解决方案，或者对于连续变量的Bayesian analysis有特殊的
方法。 |
|
c*********n
发帖数: 87 | 6 看到隔壁cowgirl的帖子，发现她是做online marketing analysis。我刚找到一
份这方面的工作，之前没有这个领域的经验，做过几个月的其它领域的
consulting。希望熟悉这个领域的各位能来聊聊！
这个公司是做online marketing的，公司就我一个做analysis的，我也是初出茅
庐。
如果能推荐一些书或者这方面的资料更好：
我现有的书是web analytics 2.0 和 web analyitics an hour a day
也在上关于marketing方面的课。
借这个帖子和各位交流交流！大家来各抒己见。 |
|
f**********t
发帖数: 1001 | 7 也顺便问一下，online marketing analysis是不是很多web data analysis啊？通常用
什么语言？ |
|
s**********n
发帖数: 111 | 8 想找银行risk analysis 的工作，版上有很多做这方面的大牛，很想学习学习，请问能
推荐些书看看吗，或者针对用logistic analysis的sas编程的案例分析之类的，有这样
的书籍吗？谢谢大牛们！！！ |
|
s**********n
发帖数: 111 | 9 想找银行risk analysis 的工作，版上有很多做这方面的大牛，很想学习学习，请问能
推荐些书看看吗，或者针对用logistic analysis的sas编程的案例分析之类的，有这样
的书籍吗？谢谢大牛们！！！ |
|
d*********d
发帖数: 169 | 10 最近经历了一个onsite面试，面完了又要提供publication samples，又说还不行，要
对他们现在的一个研究数据提出具体的统计分析办法。具体有4个问题要回答， 于是我
写了一个Statistical Analysis Plan。结果3天后说是他们决定不招人了。 我感觉受
骗了，他们是不是想骗取这个Statistical Analysis Plan啊？得到了就决定不招人了
，因为之前他们真的不知道怎么分析这个数据，有点儿复杂？ 有没有有类似经历的？
分享一下，及应对办法。多谢！ |
|
D*********2
发帖数: 535 | 11 meta-analysis no need 系统学...
Statistical Meta-Analysis with Applications |
|
s*****9
发帖数: 285 | 12 很多情况有会有一些AD HOC REQUESTS，AD HOC ANALYSIS是基于AD HOC REQUESTS 的
ANALYSIS ba. |
|
w******e
发帖数: 142 | 13 不知道"principle component analysis" by Jolliffe对你有帮助没有，这个虽然只是
factor analysis中的一种，但是这本书还是写得比较清晰明了的。不过他不是针对
sas写得。呵呵 |
|
m*********n
发帖数: 413 | 14 Bayesian analysis 没有是比较过分，难道是放在其他课，比如 advance stats 或者
什么 course sequence里面一起讲 |
|
I*******o
发帖数: 109 | 15 请问谁有这两本书的PDF:
A Second Course in Statistics:Regression Analysis 7th
Mendenhall & Sinchich
Statistical Analysis of Designed Experiments: Theory and Applications
Tamhane， 2009
我的邮箱:w******[email protected]
万分感谢！ |
|
z*****l
发帖数: 275 | 16 本学期上survival analysis，课本是John Klein的survival analysis techniques
for censored and truncated data.
请问书中提到的那些例题的data哪里能够下载到，多谢！
（未买书，下载的电子版教材） |
|
m*******1
发帖数: 855 | 17 1.cluster and segmentation 的区别
2. factor analysis and principal component analysis 的区别
多谢拉 |
|
x**g
发帖数: 46 | 18 1.cluster and segmentation 的区别
我以前做MARKET SEGMENTATION。 我们从来不用CLUSTER这个词。
2. factor analysis and principal component analysis 的区别
PCA是一种TRANSFORM。FA是对一个VRIABLE的分析，比如DESCRIPTIVE 和
对OUTLIER的处理等等。 |
|
s*****n
发帖数: 2174 | 19 1. Clusters 和 Segments 没有什么明确的定义， marketing 领域更喜欢用segments,
统计领域有专门的词汇cluster analysis
2 两者都属于multivariate analysis， 都是unsupervised, 也就是说都是只有X， 没
有Y的。
PCA 是 Z=MX
FA 是 X=MZ |
|
g*******i
发帖数: 258 | 20 The very fundamental difference is, factor analysis is a latent variable
model. The underlying factors are "hidden". Keep in mind that, in general,
when we talk about factor models, the factors are not the data. They are "
imaginary". Therefore, most textbooks will remind you the solution is not
unique.
Y = LF + E
L is the loading matrix to be estimated. F is hidden. So there will be an
identiication problem. LBB'F = LF with BB' = I. This is why any rotation of
the loading matrix can be a soluti... 阅读全帖 |
|
g*******i
发帖数: 258 | 21 Segmentation is a wider concept. It's close to "classification".
Segmentation is a purpose, and cluster analysis is one of the techniques to
achieve it.
You can use discriminant analysis, SVM, etc to achieve "segmentation". With
hierarchical Bayes methods, you can even calculate the goodness-of-fit for
EVERY sample. By ranking order the goodness-of-fits, you can also do a "
segmetation". |
|
|
P****D
发帖数: 11146 | 23 You didn't specify your hypothesis clearly so I am not sure these apply.
Just somewhere to start with.
------
Power Analysis of Non-Inferiority Tests of Correlated Proportions
Page/Date/Time 1 3/19/2013 12:43:35 PM
Numeric Results for a Non-Inferiority (One-Sided) Test of a Difference
Sample Equiv. Actual Treatment Standard Nuisance
Size Difference Difference Proportion Proportion
Parameter
Power (N) (De) (Da) (Pt) (Ps) (P01+P10) A... 阅读全帖 |
|
y*****w
发帖数: 1350 | 24 By "matched sample", you are right, paired data analysis should be applied.
However, this is still not a strictly matched sample by any means. LZ would
have to prove that the samples were matched through some mechanism (age and
gender matching, as a simple example), not arbitrarily.
LZ can check out the following article by Peter Austin. It's about
propensity score matching, with some contents on survival analysis (K-M
estimates and Cox models) in the context of matched pairs. Check out Page
113... 阅读全帖 |
|
a**j
发帖数: 60 | 25 求Solution manual for "Categorical Data Analysis" (Alan Agresti版) ，或是做
过的习题
求Solution manual for "Nonparametric statistical inference" (Gibbons版)
包子paypal， textbook pdf 交换都可以
我有SAS prep guide (base & adv)
Linear model (rencher版）
Applied Linear statistical model (Hill 版）+solution manual
Categorical Data Analysis (Alan Agresti) 2nd and 3rd editions
Statistical Inference (Cassella Berger 版）+solution manual
等等 |
|
I*****a
发帖数: 5425 | 26 hi guys, i have a question about clustering analysis with both numerical
variables and categorical(nominal) variables. I am not very familiar with
clustering analysis. Any feedback will be appreciated.Can only type chinese
using phone, which is too much pain... sorry.
1) What are the standard ways to deal with categorical variables ? Do we
simply transform them to a lot of dummy variables ? In my particular problem
, I have a pretty large dataset, where some variables may have hundred
thousands ... 阅读全帖 |
|
b*****n
发帖数: 685 | 27 啥叫PCA的power analysis？你不做test哪来的power？ |
|
j*********1
发帖数: 21 | 28 Thank you for your explanation. If I understand correctly, in a multiple
regression model, the hypothesis test is done by either T test or Chi-square
test, right? What I meant is when you are doing a power calculation to a
hypothesis test in a a multiple regression model that included one of the
predictors is a PCA index, is there anything special to it or we can just
treat it as a simple continuous variable? Sorry about the bad wording
earlier. I am not a professional statistician. :P
I think w... 阅读全帖 |
|
j*********1
发帖数: 21 | 29 哈哈，我自认小白～background就不是统计，只是半路出家修了一些课，所以
statistic theory那块都一片空白～～要是挖坑的话，我觉得此坑故事性略欠了点～～
还是回到我的问题，如果是普通regression model的power analysis还好，做个Monte
Carlo simulation就可以了。我就是concern如果有predictor是PCA做出来的index,有
没有什么东西需要特殊处理的。上网搜也搜不到什么～～呵呵 |
|
|
y**3
发帖数: 267 | 31 power analysis is to calculate sample size, is part of work of Experimental
design, as my understanding |
|
m**********4
发帖数: 774 | 32 很喜欢他那本Intro to categorical data analysis，算是我读过的最好的统计书之一
。可惜的是这书大多都是intuition，没有什么证明，听说他那个categorical data
analysis里讲的比较深，请问是吗？两本书区别大吗？挺喜欢他的书和他那篇著名的
paper的，不知道还有他的什么书推荐没有？非常感谢！ |
|
a****g
发帖数: 8131 | 33 if you are only interested in perfect match, the problem becomes a
proportion analysis
if you can expand your analysis to see agreement degree, you can try kappa
statistics
wait to see more profound insights |
|
P*****6
发帖数: 273 | 34 以前学过bayesian analysis, 用过hiarchical model.但是这次老板给了一个project
以前用的都是empirical bayes,效果还不错。我想问一下如果用真正的bayesian
analysis效果在某些情况下会好些吗？我可以试一下，但是得出的结论是个案，不具备
统计意义。所以，想请教一下大家。我感觉sample量大时，差别会较小。
另外，我体会empirical bayes有一个优点，就是速度快，如果效果差别不大的化，搞
计算机的喜欢。
谢谢 |
|
F****n
发帖数: 3271 | 35 我有3000多个individuals分布在300多个community中
Cross Sectional Analysis很简单，上multilevel就行了
但问题是我这个是20年的panel（相当于每年一个snapshot)。
有没有可以做multilevel panel analysis的tools?
Thanks! |
|
p********1
发帖数: 1011 | 36 正要用长达20年的data作分析。 我从来没做过longitudinal analysis。
请问：最好用的软件是哪个？
能不能推荐一下书：一步一步用那个软件做longitudinal analysis的书。 |
|
|
i***l
发帖数: 1656 | 38 其实我的实验Treatment是一样的，就是时间不同的几次的重复，我目前放在一个
Survival Analysis 里计算了。
3次实验都是从开始（时间为0）到最后（10-15小时），每隔1小时计数对照或样品的存
活。
因为3次重复，总体的Survival Time多少有些不同，所以我的合作者质疑，我是否应该
把3次的数据放在一个Survival Analysis 里计算 ?
我应该Pool这3次的数据在一起么？ 为什么？ 谢
model |
|
N********m
发帖数: 107 | 39 【 以下文字转载自 JobHunting 讨论区 】
发信人: Nightstorm (Swordblade,Nightstormy, Icyriver), 信区: JobHunting
标 题: Modeling/Scoring/Analysis Analyst, manager and senior manager positions
发信站: BBS 未名空间站 (Wed Apr 8 12:16:09 2015, 美东)
Consumer modeling unit in a major bank is under big expansion. There are
many positons for entry and experienced level canidates. They are hiring
Modeling/Scoring/Analysis Analyst, manager and senior managers.
Requirements: MS/PH.D. in Stats, econ or math.
Experience in using econo... 阅读全帖 |
|
a*****i
发帖数: 1045 | 40 business analysis entry level 的职位，我看他们要求挺高，需要会自己做
statistical model. 职位介绍上面主要写的statistical 只是是需要（regression ,
clustering, confidence interval, statistical significane).
部门是Financial planning and analysis,之前没学过金融经济相关的，不知到这样的
面试会遇到什么问题。他们说是需要一个统计背景的人，可以帮他们建立一下写的模型
。 |
|
a*******g
发帖数: 80 | 41 目的：比较两种不同手术方案
方法：方案一 有3个publications，每个人数在20到50不等，方案二 也有3个
publications(不同)，每个人数在10到30。outcomes都看了 术后并发症 这也是我感兴
趣的
问题：怎样把这六个single arm study整合成comparison study？
我的考虑：1，最naive的方法 把同一手术方案的人combine 一起 做chisq test
2：按照meta analysis思路 按sample size给每个study加weight 然后 combine 同一
手术方案的病人 做chisq test
这两个方案可行否 大家帮分析分析
另外还有什么办法可以做
如果按meta analysis思路 degree freedom怎么算
多谢 |
|
e****t
发帖数: 766 | 42 anybody can help download the following paper ?
Thank you in advance !! here is email: [email protected]
/* */
Baozi in return...
Some approaches to the analysis of recurrent event data.
Stat Methods Med Res. 1994;3(3):244-62.
Some approaches to the analysis of recurrent event data.
Clayton D1. |
|
B****l
发帖数: 22 | 43 I want to conduct a power analysis for correlated observations. My question is whether it is OK to treat the correlated observations at two time points (the same group of people) as independent, and conduct the corresponding power analysis for independent t-test. The reason is I want the sample size to be small, say, around 10.
Somebody told me if the sample size is large, we can treat correlated observations as independent. Otherwise, we cannot. I am not sure. Thanks. |
|
s*****o
发帖数: 48 | 44 Postdoctoral Fellowship (Medical Image Analysis and Visualization)
GE Healthcare / Western / Robarts / Lawson / LHSC
One postdoctoral fellowship is available in GE Healthcare, London, ON,
Canada.
They will conduct cutting-edge research with leading researchers in GE
Healthcare, the University of Western Ontario (Western), Robarts Research
Institute, Lawson Health Research Institute, and London Health Sciences
Centre (LHSC) in the field of medical image analysis and visualization. They
will also ... 阅读全帖 |
|
s*****o
发帖数: 48 | 45 Postdoctoral Fellowship (Medical Image Analysis and Visualization)
GE Healthcare / Western / Robarts / Lawson / LHSC
One postdoctoral fellowship is available in GE Healthcare, London, ON,
Canada.
They will conduct cutting-edge research with leading researchers in GE
Healthcare, the University of Western Ontario (Western), Robarts Research
Institute, Lawson Health Research Institute, and London Health Sciences
Centre (LHSC) in the field of medical image analysis and visualization. They
will also ... 阅读全帖 |
|
d******4
发帖数: 132 | 46 Could you give more details about how to use survival analysis on user
attrition?
thanks |
|
A*******s
发帖数: 3942 | 47 没做过marketing，
不过所有survival analysis的应用应该都差不多--
user attrition是一个event，
用一般的classification方法可以用来预测event occurrence within a certain
width of period
用survival就可以用来预测when the event will occur
logistic regression可以视为Cox model with exact/discrete time treatment of
tied events的特殊形式 |
|
c***z
发帖数: 6348 | 48 我是半瓶醋叮当晃
班上其他大牛才是深藏若虚
bayesian analysis对我来说确实是高山仰止 类似于物理里面的统一场论 |
|
P*****6
发帖数: 273 | 49 这话可以理解为bayesian 在实际中用处不大就象统一场论 ：）
bayesian analysis没有这么难，只要上课肯定能会。平时还可以用winbug.
我觉得data scientist主要工作是分析问题，找到合适的方法。大部分这些方法展现出
来后，数学推倒都不算难。关键是想透，基本功扎实不出概念错误。 |
|
P*****6
发帖数: 273 | 50 我猜您是想说paramatric survival analysis, 比如假设weibull distribution？
这个说不定对customer attrition 真的很有用。
X) |
|
