A*******s
发帖数: 3942 | 1 i would also like to know the answer. seems there is no one-sample KS test,
or other distribution tests in SAS (except for normality test).
Do u need to compare NB with other alternative distributions? Then we can
use likelihood as a criterion.
I also think such tests, if exist, lack of sense in reality. How can u tell
the prob of a sample follows NB but not Poisson, zero-inflated Poisson or
zero-inflated NB? They look quite similar...
deviance/ |
|
w*******9
发帖数: 1433 | 2 the alternative of KS test is everything other than the null, so you can
imagine that the power is very low. However, if you do like it, you can
always construct the
sampling distribution of the KS statistics under NB by simulation. It should
be very quick with nowadays computer. Based on the sampling distribution,
you can get a valid test, but again, don't have high expectation with your
power.
deviance/ |
|
f**********t
发帖数: 1001 | 3 它应该assume error项是Normal distribution的。那么它对于数据有没有这
个assumption呢？
我猜测，如果有这个assumption的话，可能是不希望有outliner。比如long
tail distribution就有很多outliner。
望指教！ |
|
f**********t
发帖数: 1001 | 4 嗯，感谢大家的回复们。
个人感觉只看到过error的Normal distribution assumption，没看到过数据的Normal
distribution assumption。呵呵 |
|
d******g
发帖数: 130 | 5 I think that's the reason why we have large number theorem and t test. The
normality assumption for error is more related to statistical inference for
beta since the distribution of beta depends on the distribution of errors. |
|
t***r
发帖数: 157 | 6 how to find the distribution of the sum of discrete and continuous uniform variable
x~uniform(0,1)
y discrete uniform 0 with p, 1 with probability 1-p
how to find the distribution of x+y?
thanks |
|
b**********0
发帖数: 53 | 7 Suppose we know the distributions of X1, X2, ..., Xn, any suggesions on how
to compute the quantiles/distributions of X1+X2+...+Xn? Note that X1, X2, ..
., and Xn may be independent or dependent; they may or may not assume the
same pdfs (or pdfs of the same form, e.g., Gaussian).
It has been shown that this computation is NP-hard in general, so I wonder
if there is any effective pseudo-polynomial algorithm or approximate
algorithms for solving the problem? Any advice will be greatly appreciated. |
|
o******e
发帖数: 1001 | 8 假定(X,Y)和(Y,Z)分别为bivariate normal distribution, 我们可以证明(X,Z)也是
bivarate normal distribution.
证明过程如下：令u,s,v,t为独立标准正态分布。
从 (X,Y) 为bivarate,我们有：
Y=a_y*u (我们省略均值)
X=a_x*u+b_x*s
从 (Y,Z) 为bivarate,我们有：
Y=c_y*v
Z=d_z*v+e_z*t
因为Y为同一变量，我们有u=v,因此，
X=a_x*u+b_x*s
Z=d_z*u+e_z*t
(X,Z)^T=(a_x,b_x,0;d_z,0,e_z)*(u;s;t)
(X,Z)是独立标准正态分布的线形转化，所以为bivarate.
但是奇怪的是，我从来没有在书上或paper上看到讨论bivarate 的transitivity,我的
证明有错吗？谢谢！ |
|
t****r
发帖数: 702 | 9 Suppose X has a normal distribution with expected value 0 and variance 1. Le
t W = 1 or -1, each with probability 1/2, and assume W is independent of X.
Let Y = WX. Then
1. X and Y are uncorrelated;
2. Both have the same normal distribution; and
3. X and Y are not independent.
Thus, (X,Y) is NOT bivariate normal.
From http://en.wikipedia.org/wiki/Normally_distributed_and_uncorrelated_does
_not_imply_independent |
|
d******e
发帖数: 7844 | 10 Covariance Matrix Rank deficient.
举个简单的例子，x服从N(0,1), y = x，那么(x,y)服从正态分布。但是这个协方差矩
阵rank只有1。无法求逆，自然也无法算density。其实Rank deficient的情况很多，常
用的PCA的假设就是covariance是low rank的。
正态分布的定义是
A random vector is said to be multivariate normally distributed if every
linear combination of its components has a univariate normal distribution.
并不是由density function来定义的。 |
|
h*****u
发帖数: 109 | 11 请教各位一个Monte Carlo在engineering里应用的一个问题。方法比较简单。是用
Monte Carlo产生若干个Sample，每一个Sample是系统的输入，然后看输出的性能。最
后统计Average。
问题是需要一个probability distribution来产生Sample。知道的信息只是包括最小值
和最大值。没有一个accepted的distribution。请问怎么办？可以把常用的Normal,
Uniform, Triangular等都用上，然后看结果的sensitivity吗？
多谢各位。 |
|
b*****e
发帖数: 2511 | 12 对一组数据进行了Lorentz fitting, 拟合的误差应该怎么算的？ Lorentz
distribution里的Gamma应该只是和这个distribution有关和拟合 误差没有任何关系吧
？ |
|
w*******9
发帖数: 1433 | 13 Otherwise how would you calculate it? The calculation based on normal
distribution is only an approximation to the true sampling distribution N(n,
p) |
|
c**********e
发帖数: 2007 | 14 Suppose I have a continuous variable X and a frequency variable Y. I fit a
model by SAS GENMOD. The model says that Y has a negative binomial
distribution with mean a+bX (or exp(a+bX)).
Specifically, I have a sequence X1, X2, ..., Xn and Y1, Y2, ..., Yn. How to
test Yi's follow negative binomial distributions with mean a+bXi (or exp(a+
bXi))?
Anybody has an idea? Thanks a ton. |
|
m******t
发帖数: 273 | 15 Yes, I have only mean and std. dev.
I only need to find shape, scale. If we can also find location, it will be
better.
I have a group of data mean and std. dev. It means that each data point in
the given set is a mean and std. dev., which are calculated from a truncated
gamma distribution.
Each point may have a different truncated gamma distribution, which has the
same truncated values.
They may have some relationship between any two data point (mean, std.dev.),
but, we do not know.
So, it makes... 阅读全帖 |
|
r*******7
发帖数: 1811 | 16 我是用copula做的，先生成multi normal的同时在var-cov matrix中把corr设置成o.5
，再转成multi unif的。
网上可以搜到怎么求joint normal distribution的公式，但找不到怎么求multi unif
的而且uniform之间是dependent的。 |
|
v*********w
发帖数: 1632 | 17 I have run a data distribution fit code below.
Can anybody help me figure out why the histogram cannot be output?
Thanks.
data Plates;
label Gap = 'Plate Gap in cm';
input Gap @@;
datalines;
0.746 0.357 0.376 0.327 0.485 1.741 0.241 0.777 0.768 0.409
0.252 0.512 0.534 1.656 0.742 0.378 0.714 1.121 0.597 0.231
0.541 0.805 0.682 0.418 0.506 0.501 0.247 0.922 0.880 0.344
0.519 1.302 0.275 0.601 0.388 0.450 0.845 0.319 0.486 0.529
1.547 0.690 0.676 0.314 ... 阅读全帖 |
|
d*********d
发帖数: 1 | 18 Just came up with this, much the same as the new approach to
share mp3 files across distributed computers ( sorry I
forgot
the name). The idea is to building a BBS that doesn't
require
a fixed server. Instead, every computer connect to the
Distributed
BBS becomes part of it. Thus, the CCP can't block the IP!
Anyone interested in this please contact me. Let us discuss
it
in detail. |
|
f****a
发帖数: 4708 | 19 50:50 wd 就是bmw自己炒作，基本就是用来恶心前驱车的。
其前置后驱被后置引擎后驱车秒杀。
911起步时，dynamic weight distribution front:rear 26:74，刹车时，42:58。
cayman 33:67 and 36:64
http://www.caranddriver.com/features/the-mid-vs-rear-engine-deb
Accurately knowing the CG location in both cars reveals that the rear (
driving) tires carry 74 percent of the 911’s weight during hard initial
acceleration versus only 67 percent of the Cayman’s. (More load equals
better launch traction.) During braking, when nearly equivalent tire loading
yields t... 阅读全帖 |
|
i***6
发帖数: 442 | 20 except for exponential, gamm, or weibul?
I am assuming the 3 lines representing the same distribution with different
parameters.
Anyone can help? |
|
l**1
发帖数: 1875 | 21 【 以下文字转载自 USANews 讨论区 】
发信人: unceasing (龙虾), 信区: USANews
标 题: If Obama wins, I will distribute 500 BAOZI on this board
发信站: BBS 未名空间站 (Tue Nov 6 18:04:35 2012, 美东)
re |
|
T*P
发帖数: 2697 | 22 re
【 以下文字转载自 USANews 讨论区 】
发信人: unceasing (龙虾), 信区: USANews
标 题: If Obama wins, I will distribute 500 BAOZI on this board
发信站: BBS 未名空间站 (Tue Nov 6 18:04:35 2012, 美东)
re |
|
z****g
发帖数: 3509 | 23 这叫屁的经济学。
你从401k ira 里拿钱也叫distribution |
|
U*E
发帖数: 3620 | 24 【 以下文字转载自 USANews 讨论区 】
发信人: unceasing (龙虾), 信区: USANews
标 题: If Obama wins, I will distribute 500 BAOZI on this board
发信站: BBS 未名空间站 (Tue Nov 6 18:04:35 2012, 美东)
re |
|
w*********s
发帖数: 2136 | 25 http://www.youtube.com/watch?v=GRwUZ-u6KFo
pakalert.wordpress.com
Former Pakistani Army General Mirza Aslam Beig claims the US Central
Intelligence Agency (CIA) has distributed 400 million dollars inside Iran to
evoke a revolution.
In a phone interview with the Pashto Radio on Monday, General Beig said that
there is undisputed intelligence proving the US interference in Iran.
“The documents prove that the CIA spent 400 million dollars inside Iran to
prop up a colorful-hollow revolution following |
|
M******l
发帖数: 479 | 26 好不容易找到一辆自己想要的配置奔驰GLC，昨天去取车的时候，dealer说manual和备
用钥匙暂时没有，说driver去distribution center的时候忘记拿了。。不知道这种情
况正常吗？一直以为manual和各种手续都应该是放在副驾的抽屉里的呢？
本来这辆车的mileage就有115迈了，dealer说是从别的地方trade过来的车所以有些里
程，然后中间放杯子的那块木板盖上还有小划痕（答应给order并且换一块），结果又
有这个小插曲，不知道这些情况正常吗？第一次买车没有经验，在怀疑是不是曾经被当
作过demo car或者被人买了退过。。dealer一直说保证绝对是全新的车。
谢谢大家！ |
|
s****n
发帖数: 1140 | 27 谁有经验，现在我正在写一个比较重要的paper，里面涉及很多计算multivariate
normal distribution，在matlab里面有mvncdf功能来实现，
如果variance-covariance矩阵是已知的，这样的计算复杂度是多少，实际上在计算站
上要消耗多少时间？我现在作3维问题，根本不算什么，如果算30维呢？哪个统计牛人
给说说
呵呵 |
|
u*******g
发帖数: 3300 | 28 【 以下文字转载自 USANews 讨论区 】
发信人: unceasing (龙虾), 信区: USANews
标 题: If Obama wins, I will distribute 500 BAOZI on this board
发信站: BBS 未名空间站 (Tue Nov 6 18:04:35 2012, 美东)
re |
|
T*P
发帖数: 2697 | 29 rew
【 以下文字转载自 USANews 讨论区 】
发信人: unceasing (龙虾), 信区: USANews
标 题: If Obama wins, I will distribute 500 BAOZI on this board
发信站: BBS 未名空间站 (Tue Nov 6 18:04:35 2012, 美东)
re |
|
s******m
发帖数: 2310 | 30 re
【 以下文字转载自 USANews 讨论区 】
发信人: unceasing (龙虾), 信区: USANews
标 题: If Obama wins, I will distribute 500 BAOZI on this board
发信站: BBS 未名空间站 (Tue Nov 6 18:04:35 2012, 美东)
re |
|
m**********w
发帖数: 4161 | 31 两段合起来看：
Distributions of Roth IRA assets from regular participant contributions and
from nontaxable conversions of Traditional IRA can be taken at anytime, tax
and penalty free
这里面提到的两个source在那个列表里是1和3，跳过了2。那不是隐含着取2的时候有可
能有tax和penalty？ |
|
h***i
发帖数: 11 | 32 看Vanguard的Distributions，里面有一项Unrealized appreciation/depreciation。
这部分是不是需要平分的税？
有的fund这部分能占30%多。是不是意味着如果买了的话，相当于用30%的钱帮前面的人
交税了（如果不巧，刚买完税就平分下来的话）
谢谢各位答疑解惑 |
|
j*****h
发帖数: 3292 | 33 401K account default distribution method will be life time annuity.
请教一下，有人知道详情么？
有没有相关的正反link
谢 |
|
c*v
发帖数: 93 | 34 显然有啊:
As of 12/07/2010 : 30-day SEC Yield 2 : 2.24%
右边还有一个Distributions的link来着. |
|
z****7
发帖数: 880 | 35 No distribution information about the fund I mentioned |
|
x*********n
发帖数: 100 | 36 不是这样的。在distribution后，股价也同时跌同样的价格。总的balance还是不变。 |
|
b******y
发帖数: 2729 | 37 【 以下文字转载自 Returnee 讨论区 】
发信人: wuliao (槑槑的槑着), 信区: Returnee
标 题: 请问腾讯的Distributed Database Engineer能有啥待遇么？
发信站: BBS 未名空间站 (Wed Mar 10 13:23:00 2010, 美东)
国内的猎头主动找到我，问我感不感兴趣。好歹我老也在美国的大公司的核心组干了2
年了，年薪也有6位数，腾讯的一个engineer position给的待遇能高到哪儿去，有人了
解么？ |
|
f*******r
发帖数: 1086 | 38 大家都如何准备large-scale, distributed system
这种面试问题的啊？有比较好的这方面的资料推荐吗？
看到说google面试会问这样的问题，觉得还是准备一下好，
请大家不吝赐教！谢谢了～ |
|
m********g
发帖数: 272 | 39 开个贴，希望大牛们讨论讨论面试中一般都会问哪些distributed system的题目。这类
题目注意一些什么？
Thanks in advance! |
|
l*****s
发帖数: 39 | 40 想海投简历了，有谁能推荐一个resume distribution的网站吗？谢谢 |
|
t*****e
发帖数: 53 | 41 I know the solution:
read the first k lines from the file,
then repeat the following steps:
- read one line from the file
- with probability X, keep the new line, and randomly drop a line from the
previous selected k lines.
- with probability (1-X), drop the new line.
till all the lines of the lines are read
My questions is:
- What should be the value of X?
- How to give a strict math proof that this method gives a randomly
uniformly distributed k lines.
thanks alot for your help. |
|
j*****y
发帖数: 1071 | 42 如何产生一个 单位圆内的 uniform distribution 呢？
多谢. |
|
l*******b
发帖数: 2586 | 43 reject sampling吧，感觉有限步uniform distribution太难 |
|
c********t
发帖数: 5706 | 44 用hash value来distribute to diff machines
如果增加一个或几个machine，该怎么办? 如果有application用这些数据，要不要
shutdown servers来操作？ |
|
l*****t
发帖数: 2019 | 45 如果不急的话，coursera上好像有门distributed system 的课。 |
|
m******t
发帖数: 273 | 46 We know the truncated point.
But, how to use the probability density function to calculate the
conditional mean and std. dev. ?
We do not know the pdf of truncated gamma distribution.
Any help would be apprecitaed.
Thanks !
mean |
|
d********m
发帖数: 101 | 47 Randomized algorithms.
Parallel algorithms.
Distributed algorithms.
NP-completeness of particular problems.
Approximation algorithms
这块有必要学吗？ |
|
X******3
发帖数: 24 | 48 百度电面谁做过啊求题目啊，c++,data structure,distribute system的题
收到HR 的电面通知求大神 |
|
发帖数: 1 | 49 【 以下文字转载自 Programming 讨论区 】
发信人: MoonChild (df), 信区: Programming
标 题: 一般distributed system用什么consistency model
发信站: BBS 未名空间站 (Tue Apr 18 17:22:40 2017, 美东)
有两个server, 互为replica, 每个都和自己的backend talk, 但database 是同一个。
要求是要implement consistency＋in-memory cache, 具体哪种consistency 也不知道
。还要fault-tolerance
有什么思路呢多谢 |
|
i*****e
发帖数: 20 | 50 你的问题不清楚。
你弄清：系统是用在什么地方，是否是地理上离很远，还是在同一个数据中心，是否两
个服务器是active active or active passive back up. Or whether there is a
primary replica and secondary replica, and replica management policy.
From request perspective, whether it's read intensive and/or write intensive
.
Once understood the above, then you can select consistency model/semantic.
1. Complete consistency. It means whenever a data item is updated, both
servers will be able to read this copy of data. In academic research, this
can be assume... 阅读全帖 |
|
