w******r
发帖数: 43 | 1 1. search literature to check the time scale for staining. I remember it
should be longer than half hour. What is the concentration of your RuO4?
2. I know some people stain the sample to the hell, then they do microtome
after
stain. The surface will basically be destroyed, but the inside of the sample
will be stained by the diffuse of RuO4.
Good luck! |
|
p******h
发帖数: 577 | 2 【 以下文字转载自 EE 讨论区 】
发信人: pegasush (飞黄－前方是转角), 信区: EE
标 题: How to get a sampled verson of laplace transformed tranfer function?
发信站: BBS 未名空间站 (Sat Jun 23 16:07:03 2007)
For example, I have a tranfer function as follows:
H(s) = (s+z1)/(s^2 + s*a + b^2);
How can I get the sampled version of it?
And what is the quite credible reference for it?
Thanks. |
|
c***g
发帖数: 357 | 3 出售PE 2008 sample questions和6 minutes solution,mechanical design部分有笔记
，HVAC和thermal部分很干净，价格站内。个人认为2008年模拟题更接近真题。同时三
本现版ncees官网卖的sample questions总共30$甩卖，PE reference manual不卖，请
勿询价。在德州和oklahoma地区Hal,SLB,NOV,BHI工作的包邮。 |
|
s******n
发帖数: 485 | 4 trough sample ---- sample collected at the end of the dosing interval? |
|
z********9
发帖数: 16 | 5 It means the smaple is collected at the end of the dosing interval, just
before the next dose is due. For example, if it is twice daily dosing, then the
trough sample will be collected at 12hr post-dose. If it is once daily
dosing, the sample will be collected at 24hr post-dose.
It is collected to indicate the drug level at that time point is within
therapeutic window.
That's my 2cents :) |
|
s**c
发帖数: 1247 | 6 create 1000 uniform random samples, one for each obs
sort by the random sample, select the first 200 |
|
l***o
发帖数: 194 | 7 Hey, can any expert give me some suggestion on the SAS code for doing the
simulation method sampling for polytomy items based on IRT method?
thanks so much.
(the main ideal is to use simulation method to create the sample that
follows certain distribution for polytomy items) |
|
j*****e
发帖数: 182 | 8 Repeatly measure a subject 10 times decreases measurement error. Varibility
from subject to subject determines if you are going to see significance, or
not. Basically, your outcome is the average of the 10 measurements from each
subject. You are comparing the mean of two samples, each with size two.
Nonparametric test will not save you. It will give you a p-value that is
larger than 2-sample t-test.
There is no majic in statistics. You have to collect information from more
subjects. |
|
s*r
发帖数: 2757 | 9 is gibbs sampling to sample from conditional distribution alternatively
noraml |
|
d*****y
发帖数: 140 | 10 有一堆样本，x_1,...,x_n,不知道啥分布
从中sample,出来的y_1,...,y_m要求服从pdf为f(x)的分布，
怎么做？可以用reject sampling么？ |
|
c**********e
发帖数: 2007 | 11 To detect a hazard ratio (HR) of 1.5 (comparing the control
group to the treatment group) (50% improvement in the median
survival time from 6 months for control group to 9 months
for the treatment group with 80% power using a 2-sided
log-rank test at the 2.5% significance level), the primary
analysis was planed to be conducted after X deaths.
How was the sample size determined? Could anybody with
oncology experience give a hint?
Some SAS sample code is as following. The thing was that
some param |
|
j*****e
发帖数: 182 | 12 The formula for this type of sample size estimation is quite complicated and
the accrual time definitely matters. First, you need to understand what
accrual time and followup time refer to. Will there be any crossover? Are
you willing to assume the time to event is exponential? What is the median
of time to event? You can download nQuery for a 30 trial. Play with
parameter setting and you will have a better feeling about all kinds of
parameter setting on the sample size.
Even after you get a sam |
|
c*****s
发帖数: 180 | 13 One possible way is generalization large sample size from this small sample
size by using bootstrap, then use the traditional way to solve it. |
|
o******6
发帖数: 538 | 14 ☆─────────────────────────────────────☆
restart (testit) 于 (Sun Mar 15 17:35:19 2009) 提到:
请问：
同一组用户，每个用户都先后用两种工具(A,B)完成同一个任务。
其中一半用户先用A完成任务, 再用B完成任务；
另外一半用户先用B完成任务, 再用A完成任务.
我的目的是比较那种工具更方便使用一些.
请问在做t-test时, 应当选用 paired samples t-test 还是
independent samples t-test.
非常感谢!
☆─────────────────────────────────────☆
lightyear11 (lightyear) 于 (Sun Mar 15 18:01:20 2009) 提到:
The respond variable is how well the target is completed(maybe time?).
I think here you have 3 predic |
|
j*****e
发帖数: 182 | 15 If you only want one sample, use proc survey select.
If you want all the possible samples, use proc plan. However, the population
has to be small for this situation and you have to do some data step to
munipulate the output. |
|
h****s
发帖数: 16779 | 16 Still do not understand what you are going to do.
Maybe you can give a makeup data example about the data structure and what
you want to test. You may do power analysis for this data first. And the
sample size calculation will easily follow.
The power and sample size analysis can be carried out only after the test
has been decided. Simply saying "need to conduct a power analysis" without
what kind of test makes no sense. |
|
w********u
发帖数: 328 | 17 I want to calculate the POWER.
SAS code:
proc power;
onesamplemeans
mean = 8
ntotal = 150
stddev = 40
power = .;
run;
This method makes sense when it follows normal dist., but my case is that
the sample size is extremely small but the population is very large, i can't
say that it follows normal dist., i assume it doesn't follow any particular
dist. but i haven't got the data yet, i wonder which method could be used
in this case and what's the sample size needed to conduct any analysis. for
examp |
|
j*****e
发帖数: 182 | 18 How did you sample from the pop with all ages?
What about the sample size of elderly?
Randomized means iid. Are subjects iid here? |
|
p***r
发帖数: 920 | 19 there is no sample is randomized, every sample is fixed, after you stick
with a design. |
|
l*******n
发帖数: 19 | 20 "Random sample" and "Randomized design" are different concepts.
If there is no treatment intervention, you have an observational study.
Your samples (no matter old or not) are still random.
If there is treatment intervention, the subgroup design (old people) may
not be randomized even your overall design is randomized, when your design
was not stratified by age (with old as a stratum). |
|
a****r
发帖数: 1486 | 21 你的意思是不是sample beta from p(beta|y)？？
你title上不是都说了，是用的gibbs sampling吗？？
p |
|
x***x
发帖数: 3401 | 22 sample survey主要是知道什么情况下用什么公式 其实不难
搞清楚几种sample方法就不难了 公式看着复杂但不要被吓到 |
|
s******n
发帖数: 134 | 23 Thanks very much for the reply. I really appreciate it. How to obtain a
random sample of size 1000 from f(x,y)=c? I know how to generate random
sample from normal distribution,dirichlet distribution,and so on. Thanks a
lot~ |
|
E*****p
发帖数: 39 | 24 目前正在找工作的初级阶段，有两个公司让我发了SAS code sample
我的code不是很长（1-2页），不过也有MACRO ODS，自己觉得还算系统
可是发出去然后就没有动静了，不知道像这样的情况，一般要什么样的sample code呢
？ |
|
l*****i
发帖数: 43 | 25 i am a new comer, my boss asked me to get the sample size for a clinical
trial, however, using my way, i can not get the number the former
statistician got. anyone has an idea how she got it?
1. alpha=0.05
2. power=80%
3. my boss wants 10% drug effect (she didn't explain in detail what this
mean, i assume it is after medication, the score drops 10% more than control
, because the lower the score, the better the treatment).
4. population mean drop is -0.17 (it is a score to evaluate mental functi... 阅读全帖 |
|
d******o
发帖数: 59 | 26 I think you should delete them, espeically,when the sample size is big.
When the sample size is small, it's harder to find outliers, since the
variance is big. |
|
s*********t
发帖数: 173 | 27 看到一个比较对口的工作，要求交一个writing sample. writing sample 是一个比较
专业的东西，本人只是个小硕，没有以前保留过的好文章。现在让我写，似乎很难写出
质量好的东西来，对硕士学的东西也处于一知半解中。该怎么办呢？ |
|
j*********n
发帖数: 5 | 28 一个看似很简单的问题，但是不知道怎样入手：
population由单个个体组成，每个个体可能有错误，怎样决被检查sample的size才能在
统计意义上sample的错误率代表population的错误率是significant呢？margin of
error怎样求？ |
|
c****g
发帖数: 156 | 29 adjust the sample size by taking the sampling variability into account. |
|
c****s
发帖数: 395 | 30 logistics regression
i tried using both whole data and a sample.
the interesting is that using whole data some levels which are not
significant under small sample becomes highly significant . |
|
A*******s
发帖数: 3942 | 31 the most intuitive reason is that u happen to sample a subset in which those
variables are significant.
my further guess is smaller sample size brings in collinearity and it then
causes changes in coefficient estimates. |
|
z**********i
发帖数: 12276 | 32 一个CLINICAL TRIAL的问题.一种新的癌症治疗需要多少病人?
计算新方法每年的有效率,报道的常规方法有效率.
预期:新方法有效率-已有方法有效率>=10%
how many sample size do I need with one-sample test, 10% absolute
improvement, 90% power and one-sided a=0.10.
Thanks. |
|
g******n
发帖数: 339 | 33 你这个自我称呼让我回到了遥远而性感的古代。“垆边人似月，皓腕凝霜雪”，哈哈，
想得太多了。
回到这个讨论的问题吧， 一般来说，effect size 指的是diffrence of the means or
proportions。不过在medical jurnal上好像也有人把这个均值的差除以一个standard
deviation,算是一个standardized effect size. If the end point is a
continuous variable, and the comparison is between 2 groups with common
standard deviation assumption, then the standardized effect size, along with
alpha and beta, would be sufficient for a sample size calcultion based on
normal approximation. However, this is not the case for ... 阅读全帖 |
|
s*****r
发帖数: 790 | 34 你这个问题显然不满足正态的假设。你是想估计proportion.
通常情况下 exact 95% CI 不会是对称的（一般大家求的，不是说不能,symmetric if using normal approximation）。你还需要有一个假
设，就是真实的rate是多少。rate越接近0.5 sample size 就需要越大。
I suggest the following sample size calculator. (ps I am not familiar with survey data so the above is just general statistical consideration)
http://www.raosoft.com/samplesize.html
个数除以广告展现数）。这个rate在我总
normal distribution, 95％ confidence
formula |
|
d*********k
发帖数: 1239 | 35 2 sample t-test with bootstrap怎么弄啊? 第一次用bootstrap没有什么经验 :(
有两组数据，sample size都比较小，一个8个，一个13个，现在想用T test with
bootstrap比
较它们的mean，怎么弄啊? R里面有相应的package么？
数据A： a1 a2 a3 ......... a13
数据B： b1 b2 b3 ... b8
怎么做呢？谢谢啦啊~~ |
|
S********a
发帖数: 359 | 36 请教个sample size calculation for logistic reg的问题，predictor是PM(25%的人
是在exposure下，trial的个数我自己写了个1000）, outcome 是MI(4个人里有一个人
是MI，所以P＝0.25,trial的个数也是自己写的1000),我想看sample size 要多少才可
以significantly odd ratio>1.
proc power;
logistic
vardist("pm")=binomial(0.25, 1000)
vardist("MI")=binomial(0.25, 1000)
testpredictor="MI"
responseprob=0.25
testoddsratio=1
alpha=0.05
power=0.8
ntotal=.;
plot y=power min=0.01 max=0.99 yopts=(ref=0.8 crossref=yes) vary(symbol);
run;
结... 阅读全帖 |
|
R*******c
发帖数: 249 | 37 一个模型，用了mcmc去估计模型参数，是gibbs sampling和metropolis-hasting混着的，
现在想证明这个算法的一些性质，不知道一般如何证明MCMC算法得到的sample就是来自
targeted distribution的呢？
没有头绪啊，如果有些入门的paper或者材料读读就好了
多谢 |
|
G*****t
发帖数: 57 | 38 Hi, all,
I have a question about calculations for two independent sample...
Suppose two independent sample (A, B), only CV for each is given (CV_A, CV_B
), df also given (dfA, dfB). And we want the CV for the total population...
Here's what I thought:
Var(A+B) = VAR(A)+VAR(B)
Assume mean_A = mean_B,
CV(A+B) = sqrt(CV(A)^2+CV(B)^2)
Is there any problem for this calculation?
Shall I consider df values too as a weight? Say,
CV(A+B) = sqrt((dfA*CV(A)^2+dfB*CV(B)^2)/(dfA+dfB))
Thanks a lot!! |
|
|
l****p
发帖数: 27354 | 40 【 以下文字转载自 JobHunting 讨论区 】
发信人: lulupp (木有昵称), 信区: JobHunting
标 题: 借人气问一个Excel sampling 问题
发信站: BBS 未名空间站 (Thu Feb 16 15:59:21 2012, 美东)
假定Excel格子里（A1:A9）有9个数：1，2，..., 9.
现在我要随机抽取5个，每次抽一个，抽中的数字不能重复（=sampling without
replacement）。每次抽的时候，要求剩余的数被抽中的几率和数本身的大小成正比。
这样5个数一组，我要很多组（比如5000组）。
谢谢。 |
|
p***r
发帖数: 920 | 41 看了很多 textbook 和资料，基本上 PPS 是一种 cluster sampling with unequal
probability. clustering sampling 需要有 primary unit 和 secondary unit(
subunit, element).
我现在的 project 要做一个 survey, 但是只有一种 unit（房子）和它的某种意义上
的 size（用电量），不能找到明确的 secondary unit 来定义（因为一个房子就对应
一个survey report result）。
请问这种情况下能使用 PPS 么？ 我看了一下书，书本上在定义上提到了secondary
unit, 但是在计算 statistics 的时候都没有用到过。 |
|
w*****t
发帖数: 49 | 42 有一个大的population，大概4million，想从里面选20k出来。need control variable
"age"
population的mean（population age）=50
选出来的sample希望 mean（sample age）=20
请高手帮忙！3个包子奉上。 |
|
m**c
发帖数: 199 | 43 怎么样sample从下面的distribution
1. wN(0,1)+(1-w)N(1,1)
2. p(y|x)=C*p(x|y)p(y), 可以assume p(x|y)=N(y,1) and p(y)=N(0,1)
直接sample，不要先算distribution
谢谢！ |
|
m**c
发帖数: 199 | 44 metropolis-hasting应该可以。
可以先sample y~p(y),然后用 p(x|y)作为 weight,重新sample吗？ |
|
a********y
发帖数: 474 | 45 suppose it's a survey data with sampling weight
treatment: A (0,1) ,B (0,1)
sex: male; female;
age groups: young, middle, old
all patients may use A, or B, or both A and B at the same time. So the
independent sample assumption does not hold.
Question: is treatment A significantly more adopted than treatment B (
overall, stratified by sex, age, etc)?
请问可以 dependent t-test 吗？
谢谢！！ |
|
w*******9
发帖数: 1433 | 46 paired data variability 小所以需要的size小，这是为什么要pair. 比如x=rnorm(
100), y=x+0.01, 如果是paired, pvalue=0. 如果是unpaired, pvalue会很大。
你说的paired survival sample size 很有意思，你看见的sample size
modification 能给个链接吗
？我想学习学习。
),
design |
|
i****f
发帖数: 979 | 47 请问一下如果sample是从不同的 normal distribution 里抽来的，硬要做one-sample
t-test的话，后果是什么？ |
|
K*******5
发帖数: 98 | 48 Thank you for your help.
It looks like i did not make my question clear. I need random samples from
CONTROLS. The samples from CONTROLS need to match each observation from
CASES by site and year. The ratio is 5:1. 5 controls to 1 case. |
|
|
c********h
发帖数: 330 | 50 Yes.
You first want the margin error. It's 0.01 for your case?
You also want a confidence level, say 0.95
You also want to know what statistic is interested in. I'm not quite sure
what you want to use, so let's just suppose it's some sample mean. The most
commonly used statistics in this situation is either a proportion or a
sample mean.
Then margin error = t(alpha) * s/sqrt(n),
where t(alpha) is the upper quantile of t-dist. s is the standard deviation
for your data, either know or use bootstra... 阅读全帖 |
|
