|
y*********s
发帖数: 24 | 2 希望高人指点:
做MULTI TESTS,需要得出ADJUSTED P-VALUES, 想用FDR PROCEDURE(B-H algorithm).
R 里有两个PACKAGES, multtest and FDR-AME, 可以做.
问题是:
1 在INSTALL LIST 里(R2.11.1)找不到FDR-AME;
2 DOWNLOAD multtest后, R 需要Biobase? 又找不到Biobase;
3 用这两个PACKAGES 得出的结论不太一样, 而它们用的方法是一样的,
都是B-H algorithm, 该怎么看呢?
4 若有example codes 就更好了;
谢谢! |
|
e*******e
发帖数: 2 | 3 There are currently two job openings in Covance (www.covance.com).
Good luck!
Senior Biostatistician I
Master’s degree in statistics/Biostatistics/Applied statistics with
excellent programming skill in SAS, particularly expertise in IML, Proc GLM
and Proc Mixed, Proc MULTTEST, Proc GENMOD, and power/sample size
calculation of various experimental designs and Macros. Major emphasis will
be on carcinogenicity testing statistics and safety pharmacology (with
Latin Squares with embedded repeated me... 阅读全帖 |
|
r****q
发帖数: 22 | 4 True. It requires packages ‘graph’ and ‘multtest’to be installed. But
you may install them without Bioconductor too. |
|
y***j
发帖数: 11235 | 5 真集是啥东西？google了一下都是美女写真集。
你说的情况，不就是审稿的觉得用false discovery rate比较好，你非要用没有矫正过
的p值么。跟你说的每个人一个标准一个意思么。没有矫正的过的pvalue肯定不十分适用于multtest，原因我上面已经说过了
。或者你不怕假阳性，这样搞也可以，后续验证的工作量等着你。
而且为什么不能同时用调整后pvalue（当然这里有不同方法选择的问题）和fold change（有几倍选择的问题），但是跟没调整过的p值来比，已经不是核心问题了。 |
|
|
j*****e
发帖数: 182 | 7 This will do.
proc surveyselect data=set
method=srs sampsize=2
rep=5000
seed=40070 out=SampleRep;
run;
You can also use Proc multtest. |
|
r****q
发帖数: 22 | 8 Forgot to mention, this gage package requires packages ‘graph’ and ‘
multtest’to be installed too. But you may install all of them individually
without Bioconductor. |
|
Y******Y
发帖数: 8753 | 9 我说的这些R都有现成package做的。
multtest, qvalue
基本上就是你通过hypothesis testing得到一堆pvalues之后，用一些方法（包括
permutation tests）重新搞出一些adjusted的p-values来控制family wise error
rate就是我们平常说的type I error rate.
而storey提出的qvalues是针对false discovery rate来做的,less conservative. |
|
c****y
发帖数: 94 | 10 It sounds like that you did a pairwise comparisons among those methods. You
need to apply adjustment. You actually should have 10 pvalue (5x4/2). If you
want to use Bonferroni Correction, it may be too conservative.In sas, you
can use proc multtest with hommel option (different way to control family
error rate). Hope it can work for you. If adjusted p-value is not
significant, it just means there are not significant difference between the
methods (2 methods compared). Actually, you should do ANO... 阅读全帖 |
|
s*********a
发帖数: 2623 | 11 Hi，小辈（新手）请教大侠们：
我有4个within subject 和4个between subject的data。在这上面外带了一个Time
variable。
我想要找出在某些时间点with-in，between subject，以及correlation是否
significance different from other time。
介入data很大。是在iml下面做好还是multtest下permutation做好？是不是iml下面的
running time 更短呢？还有就是time variable,不知道要怎么处理比较好。。。还是
说我思路不太对直接上mixed/glimmix?
谢谢指教 |
|
