y****d
发帖数: 432 | 1 ★★★★★★★★★★★★★★★★★★★★★★★★★★★★★
前面说明：
需要的童鞋请到我的签名档的博客查找！谢谢！发E-mail太累了！
觉得有价值的话可以顶一下，以便更多的人看到！谢谢！
★★★★★★★★★★★★★★★★★★★★★★★★★★★★★
本合集包含内容（总有你想要的吧？！呵呵）：
A Basic Course in Probability Theory
A Course In Probability Theory.djvu
A First Course in Statistics for Signal Analysis
A Handbook of Statistical Analyses Using R
A Handbook of Statistical Analyses using SPSS
An Introduction to Probability and Random Processes
An Introduction to Probability and Statistical Inference
An Introduction to Probability Th... 阅读全帖 |
|
y****d
发帖数: 432 | 2 ★★★★★★★★★★★★★★★★★★★★★★★★★★★★★
前面说明：
需要的童鞋请到我的签名档的博客查找！谢谢！发E-mail太累了！
觉得有价值的话可以顶一下，以便更多的人看到！谢谢！
★★★★★★★★★★★★★★★★★★★★★★★★★★★★★
本合集包含内容（总有你想要的吧？！呵呵）：
A Basic Course in Probability Theory
A Course In Probability Theory.djvu
A First Course in Statistics for Signal Analysis
A Handbook of Statistical Analyses Using R
A Handbook of Statistical Analyses using SPSS
An Introduction to Probability and Random Processes
An Introduction to Probability and Statistical Inference
An Introduction to Probability Th... 阅读全帖 |
|
A**H
发帖数: 4797 | 3 【 以下文字转载自 Biology 讨论区 】
发信人: APHH (hutu), 信区: Biology
标 题: 一个统计问题求助，z score and probability
发信站: BBS 未名空间站 (Tue Oct 13 19:26:25 2015, 美东)
在看一篇文章
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2752127/
里面方法部分有一句话，不是很明了:
we first convert the read count of a window into a Z-score zi by subtracting
the mean of all windows and dividing by the standard deviation. The Z-score
is then converted to its upper-tail probability p_Upperi = P(Z>zi), and its
lower-tail probability p_Loweri = P(Z
这里说把z score... 阅读全帖 |
|
A**H
发帖数: 4797 | 4 在看一篇文章
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2752127/
里面方法部分有一句话，不是很明了:
we first convert the read count of a window into a Z-score zi by subtracting
the mean of all windows and dividing by the standard deviation. The Z-score
is then converted to its upper-tail probability p_Upperi = P(Z>zi), and its
lower-tail probability p_Loweri = P(Z
这里说把z score转化成probability，搞不懂。我知道怎么把z score转化成
percentage，因为+/- z score到0之间的面积百分比在 normal distribution里面有表
格可以转换。但是它这里的upper and lower probability到底是个... 阅读全帖 |
|
m*****e
发帖数: 268 | 5 下面这些期刊如何，大概能分成哪几档，谢谢
Annals of Probability
Probability Theory and Related Fields
Annals of Applied Probability
Journal of Applied Probability
Advances in Applied Probability
Stochastic Processes and Their Applications
Stochastic Systems
Stochastic Models
Queueing Systems |
|
l**********4
发帖数: 27 | 6 annals of probability 和 Probability Theory and Related Fields一个档次， 都
是概率论的最好的期刊。
Annals of Applied Probability 和 Stochastic Processes and Their
Applications 都是应用概率最好的期刊。差不多。（applied可能好那么一点点。）
annals of probability 和 Annals of Applied Probability 不好比较，因为focus不
一样。Annals of Applied Proba上经常也有神作。
其他的不清楚。 |
|
m****r
发帖数: 141 | 7 【 以下文字转载自 Quant 讨论区 】
发信人: mitcar (mitcar), 信区: Quant
标 题: An interview question about probability.
发信站: BBS 未名空间站 (Sat Mar 17 21:13:44 2012, 美东)
An interview question about probability.
You enter a stadium with 1000 seats. you are told that under one of the
chairs is a prize. You choose a seat randomly.
q1, what's the probability your seat has a prize
q2, now 990 seats are removed, not including your seat or the one with the
prize under it. There are 10 seats left. What's the probability that ... 阅读全帖 |
|
b*******d
发帖数: 32 | 8 It is t/2.
Because the distribution is uniform, let's consider the express train which
arrives every 6 minutes. The probability to wait time t is dt/6, therefore,
E(t)=\int_0^6 t dt/6= 3.
My computation for the problem is as follows:
For the express train (arrives every 6 min), the probability to wait time t is
dt/6, denoted as A
for the regular train (arrives every 3 min), the probability to wait time t is
dt/3 denoted as B
The probability to catch express OR regular train after time t is:
P(A |
|
s*****e
发帖数: 115 | 9 Sorry, please ignore my previous post. I think I agree with your answer now.
Here's how I got it:
Person 1 choose a number from 1 to 100 according to the following the
probabilities:
p1, p2, ..., p100
Person 2 guess a number from 1 to 100 according to the following the
probabilities:
q1, q2, ..., q100
The expected payoff to Person 2 is:
E=p1*q1*1+p2*q2*2+p3*q3*3+...+p100*q100*100
Person 1 wants to minimize E while Person 2 tries to maximize E, both
subject to the constraints:
p1+p2+...+p100=1
q1... 阅读全帖 |
|
i******n
发帖数: 839 | 10 Found a good review instead of any download link:
Summary: The truth about this book. 99% of other great reviews are bogus.
Rating: 2
This book was not written for students. It was written so that the author
can gain respect from his from his academic peers. The explanations are
absolutely horrible. It purposely explains simple concepts in overly verbose
, complicated ways. The idea is to make the subject appear as complicated as
possible when it doesn't need to be. It reads like those academic ... 阅读全帖 |
|
m****d
发帖数: 372 | 11 May 9 (Bloomberg) -- Sales at U.S. retailers probably rose in April for a
seventh straight month, pointing to a rebound in consumer spending that is b
roadening the recovery, economists said before reports this week.
Purchases increased 0.2 percent in April, extending the most successive gain
s since 1999, according to the median estimate of 60 economists surveyed by
Bloomberg News before Commerce Department figures on May 14. Other reports m
ay show manufacturing picked up and the trade gap wa... 阅读全帖 |
|
b***y
发帖数: 14281 | 12 possibility 是指有可能还是没可能，更注重可能性是否存在，probability是概率，
侧重可能性到底有多大，简单说就是
possibility <=> probability >0
no possibility <=> probability =0 |
|
|
a*****y
发帖数: 613 | 14 就是more than 50% 可能性的意思
比如50.1% probable,49.9% not probable，就可以说more probable than not |
|
c*********n
发帖数: 128 | 15 A graph {E, V}. The edges of the graph are assigned randomly between
vertices.
For two random vertices i and j, with degrees k_i and k_j, respectively,
what is the probablity that there is an edge between them?
I think the result is
k_i*k_j/m
where m is the total number of edges in the graph.
For any edge E, the probability that E is connecting vertices j is
k_j/m
Considering i is connected with k_i edges, then the probability is
k_i * (k_j/m)
But some paper gives
k_i*k_j/(2m)
What do you guys s |
|
m********e
发帖数: 16 | 16 to learn probability theory, there are two ways, one is the very formal way by
study measure theory. probability is just a special kinda measure function, w/
all the properties you might have been familiar with. the other is the
traditional way taught in your college, like most of the introductory books.
measure theory is basic tool for real analysis.
anyway, after you have learnt these stuffs, you will have a very keen view, if
you learn it well, on probability theory.
if it is just a statistic |
|
l****y
发帖数: 92 | 17 【 以下文字转载自 Quant 讨论区 】
发信人: leephy (leephy), 信区: Quant
标 题: two probability problems?
发信站: BBS 未名空间站 (Mon Feb 18 21:12:43 2008)
1) suppose U is a continuous uniform [0,1] random variable. what is the
probability that the decimal expansion of U contains no fives?
2) a stock is currently worth $100, assume that each of the next 10 days the
stock either increases or decreases in value by $1. what is the probability
that over the next 10 days, the stock reaches a maximum value of $104 and
on day 10 |
|
B*********h
发帖数: 800 | 18 ☆─────────────────────────────────────☆
chopinor (lonelycat) 于 (Wed Dec 27 16:30:21 2006) 提到:
first, sorry I misunderstood the previous problem by bigbendan,
the solution could be derived in a straighforward way
However, from my misunderstanding I have a more interesting problem to share
now :)
Consider you are just tossing a fair coin,
(1) what is probability of getting TH ahead of HT?
(2) what is probability of getting TH ahead of TT?
(3) In general, what is the probability of getting som |
|
B*********h
发帖数: 800 | 19 ☆─────────────────────────────────────☆
mymiracle (miracle) 于 (Thu Apr 19 18:49:09 2007) 提到:
You have $100 initially. You are playing a repeated game with a guy with an
infinite amount of money. You have a 51% probability to win each game and 49
% probability to lose each geam.
Each time you earn $1 if you win, lose $1 if you lose the game. What is the
probability that you will eventually go broke?
☆─────────────────────────────────────☆
Chase (大通) 于 (Thu Apr 19 18:57:59 2007) 提到:
1？
|
|
s****i
发帖数: 216 | 20 Two cases,
one is that you memorise the result, once you crack the 1st bit, you only
have to deal with 2nd bit.
The expected number is E[x] = (1+2+3+4+5)/5 + 2*((1+2+3+4+5)/5)- 1 = 8
The other is just randomly choosing 1 from 5.
let p1(k) be the probability that succeed in round k,
p2(k) be the probability that pass 1st bit in round k,
p3(k) be the probability that fail in round k
suppose you randomly pick from the 5 numbers.
p1(k) = p2(k-1)* 1/5
p2(k) = p3(k-1) * 1/5
p3(k) = p2(k-1)*4/5 + p3(k- |
|
f***a
发帖数: 329 | 21 my solution:
The task can be divided into two step:
step 1: get the 1st digit correctly (light goes on)
step 2: get the 2nd digit correctly
let n1: number of keypresses to complete step 1
n2: number of keypresses to complete step 2
n1 can be 1,2,..,5 with corresponding probability 1/5 for each
n2 can be 1,2,3,4 with corresponding probability 1/4 for each (because no
repeats)
thus n=n1+n2 and n can be 2,...,9 with corresponding probability listed
above.
You can draw a tree to understand this |
|
m****r
发帖数: 141 | 22 This is an interview question of probability. The interview has been done.
In a room, there are 8 drawers in a desk, a man put a letter into one of
them with 50% randomly. After that, you enter the room, if drawers from 1 to
7 have been opened to show they are empty, what is probability of the 8th
drawer has a letter ?
My idea:
this is a conditional probability question and can be solved by Bayes
formula.
But, I got result 14/29 different from interviewer's answer.
Any help will be appreciated.
... 阅读全帖 |
|
m****r
发帖数: 141 | 23 An interview question about probability.
You enter a stadium with 1000 seats. you are told that under one of the
chairs is a prize. You choose a seat randomly.
q1, what's the probability your seat has a prize
q2, now 990 seats are removed, not including your seat or the one with the
prize under it. There are 10 seats left. What's the probability that your
seat contains the prize now?
My answer to q1: 1/1000
Q2:
A: my seat has a price , B: 990 empty
P(A|B) = P(B|A)P(A) / ( P(B|A)P(A) + P(B|A^c)... 阅读全帖 |
|
t***r
发帖数: 3 | 24 Suppose m balls are put into n boxes (m>=n) such that each
ball is put with equal probaility (i.e. 1/n) independently
of one another into one of the boxes.
What is the probability that each box contains at least one
ball, i.e the probability of no empty boxes? If there is no
single-term expression, what is the best way to compute the
probability in terms of numerical accuracy? (For example,
use of inclusion-exclusion formula is highly undersirable in
this respect, since it contains substractions |
|
d*z
发帖数: 150 | 25
1.The probability that the first boxes without balls is
(1-1/n)^m,
and some is the probability that the second boxes and etc.
2.The probability that the first two boxes without balls at
the same time
is (1-2/n)^m.
.....
So the result is
1-C(n,1)(1-1/n)^m + C(n,2) (1-2/n)^m-...+(-1)^k
(1-k/n)^m+...+(-1)^(n-1) (1/n)^m |
|
a***r
发帖数: 420 | 26 =====ZZ
denotation:
O={O1,O2,O3...OT} observation
Q={q1,q2,q3...qT} inital state
S={1,2,3...S} possible states
λ={A,B,pai}
*A:transition probability alpha(a); B: emission probability beta; pai:
initial distribution
##不懂的部分：##
beta t(i)=P(Ot+1,Ot+2,...OT|qt=Si,λ)
1)initialization:
beta T(i)=1, 1<=i<=N
2)induction:
beta t(i)=Σ(j=1~N)aijbj(Ot+1)bt+1(j),
t=T-1, T-2,...1, 1<=i<=N.
=====ZZ
会者不难，以上是一本教材里用forward-backward algorithm解HMM中backward
probability的推导，希望脚标表达还算清楚（不清楚我再贴截图）
实在是看得困惑啊
第一步 |
|
D****G
发帖数: 284 | 27 一个loan default 的data sample, sample size > 50,000, true default rate 2.5%
, 用SAS logistic 里的OUTPUT statement和Prodicted= 得到的Estimated
Probability of Default即使对于真正的Default Event来说都很小(<20%)。而且 Non-
default event的Estimated PD 与default event 的EstimatedPD有差别但很小。算出
来的所有Estimated Probability of Default都小于25%
理想的状态不是应该大部分default event的Estimated Probability of Default都很
高么？
也用over-sampling试过，做过intercept adjustment,结果差不多。 |
|
C***r
发帖数: 759 | 28 Causality and Probability
It is sometimes argued that, in tune with the German
idealistic tradition, Weber rejected the notion of causality
in human affairs. This is emphatically not the case.
Weber firmly believed in both historical and sociological
causality, but--and this may have given rise to
misunderstandings--he expressed causality in terms of
probability. Such stress on chance or probability, however,
has nothing to do with an insistence on free will or the
unpredictability of human beha |
|
D**S
发帖数: 24887 | 29 Probable is somewhat more possible than possible.
Likely is more possible than probable. |
|
x****u
发帖数: 12955 | 30
Probably <> Probability
Possibly <> Possibility |
|
c*******o
发帖数: 8869 | 31 next time your boss say to you "I will probably fire you". Please feel good
about yourself as he just "probably"... hehe. |
|
a******g
发帖数: 318 | 32 【 以下文字转载自 ebiz 讨论区 】
发信人: asihuang (asihuang), 信区: ebiz
标 题: 10个包子弱弱问 这个概率密度函数(Probability density function)怎么求
发信站: BBS 未名空间站 (Mon Aug 4 18:18:55 2014, 美东)
实在不好意思问老板，来这里高手如云的地方问问？
假设有一个连续函数h：
h(x,y)=x^2+y^2+2x,
其中x=[0 1],y=[0,1].
(其实h是surface height，当然，它是什么不重要）
请问一下h的概率密度函数（Probability density function）怎么求？
我想知道理论解(closed-form expression)怎么求，不要让我用matlab离散方法做。
10包子准备奉上给第一个解答的，谢谢。 |
|
b********y
发帖数: 5829 | 33 Jan. 17 (Bloomberg) -- The index of leading indicators probably rose in
December for a ninth month, while home construction was little changed,
indicating housing’s role in the U.S. expansion is waning, economists said
before reports this week.
The Conference Board’s measure of the outlook for the next three to six
months probably climbed 0.7 percent last month, according to the median
forecast of 41 economists surveyed by Bloomberg News before the research
group’s report Jan 21. The Commerce De |
|
j***f
发帖数: 3610 | 34 On September 29th, digicame-info published those Nikon D800 specs
(Google translation):
What number of pixels is 3630 megapixels.
4 frames per second continuous shooting in body only, about 6 frames per
second will be used to DX mode with the optional battery pack.
A little late to be released by the model resolution sought to eliminate
the low-pass filter.
Full HD video in 1920 × 1080/30p.
Headphone jacks, can be input from an external device such as a PCM
sound recorder. Corresponding to USB3.... 阅读全帖 |
|
f******h
发帖数: 269 | 35 最近偶然看到Gregory,Porter在今年的极地音乐颁奖典礼上面唱Sting的It＇s
Probably Me.。
Gregory,Porter是我非常喜欢的一位爵士歌手，他的两张专辑《Liquid Spirit》和《
Take Me to the Alley》在2014年和2017年被评为格莱美年度最佳人声爵士专辑，但是
我不得不说，Gregory,Porter这一版本的It＇s Probably Me.改编得不好。
我记得第一次听这首歌的时候，是作为一部电影的插曲出现的。当时给我的感觉就是和
Sting的那首传唱度很高的Shape of my hert很像，但是这两首歌的写作手法不同，前
者更多的是隐喻，后者大概是白描和直抒胸臆。当时在电影中，这首歌想要表达的是手
足之情，但是现在我更愿意换另外一个角度去解读这首歌。原创版的那个Me指的是手足
，现在我愿意把它理解为自己。因为友情这个东西，其实在某种程度上和爱情和亲情一
样，不是每个人都可以那么幸运，都可以拥有的。
听了很多遍这首歌，我觉得这首歌的主要亮点有三个。一个是作为引子的打火机，一开
始听这首歌的时候，我很纳闷曲子开头几... 阅读全帖 |
|
C***r
发帖数: 759 | 36 【 以下文字转载自 ZST 讨论区,原文如下 】
发信人: Camer (铁划银钩), 信区: ZST
标 题: Causality and Probability
发信站: The unknown SPACE (Mon Sep 25 15:50:15 2000) WWW-POST
Causality and Probability
It is sometimes argued that, in tune with the German
idealistic tradition, Weber rejected the notion of causality
in human affairs. This is emphatically not the case.
Weber firmly believed in both historical and sociological
causality, but--and this may have given rise to
misunderstandings--he expressed causality in terms of
pro |
|
f***h
发帖数: 52 | 37 【 以下文字转载自 Mathematics 讨论区 】
发信人: fisch (-_-), 信区: Mathematics
标 题: 请教: probability mass function的中文是什么?
发信站: BBS 未名空间站 (Wed Jul 20 02:08:13 2005), 站内
发信人: fisch (-_-), 信区: Statistics
标 题: 请教: probability mass function的中文是什么?
发信站: BBS 未名空间站 (Wed Jul 20 02:03:23 2005), 站内
谢谢 |
|
|
s******s
发帖数: 13035 | 39 人家定义不是都写出来了
upper-tail probability p_Upperi = P(Z>zi)
lower-tail probability p_Loweri = P(Z
subtracting
score
its |
|
f******r
发帖数: 2975 | 40 Hi I have a Markov Chain with transition matrix P
starting from x_0, we pass x_{j1}, and reach x_{j2}
now I am trying to derive the probability
P(x_0 -> x_{j1} ->x_{j2})
here is my way.
Suppose initial probability for x_0 is Q
it is a nx1 vector
P(x_0->x_{j1})=QP^{j1-1}P_{x_j1}P^{j2-j1-1}P_{x_j2}
I don't like the two vectors here P_{x_j1} and P_{x_j2}
which stands for the x_j1-th column of matrix P
Any other ways to do that?
thanks a lot! |
|
B*M
发帖数: 1340 | 41 那你说该怎么办？
I am sure that a separable Banach space can be a probability space,
so I believe a probability space must be a Banach space,
but I don't know how to prove it. |
|
e**********n
发帖数: 359 | 42 Re 上十大，我來助一臂之力。
BTW, Like someone said above, define a norm with the special probability
measure or whatever other
structures you have on the probability space, and prove the completeness.
That's it. In general, this is not
a valid proposition. |
|
v*********y
发帖数: 667 | 43 我以前不是学数学的，我已经12年没有碰数学了，大学以前数学还拔尖，如果不和后来
进入数学专业的人比。Calulus的知识应该够用。现在想学一下mathematical
statistics or probability，要求taught using a calculus model, 难度和
Mathematics, statistics, physics, or engineering majors的相当。请教大家，有
什么比较好的教材？
大家知道纽约有什么学校开设交钱就可以修课的mathematical statistics or
probability课程？
谢谢了。 |
|
a*****n
发帖数: 5158 | 44 【 以下文字转载自 Statistics 讨论区 】
发信人: ashdown (EB1A and NIW), 信区: Statistics
标 题: 问一个线性regression的probability of fit怎么算
发信站: BBS 未名空间站 (Sun Oct 4 18:44:40 2009, 美东)
有一组数据
X(i),error_X(i), Y(i),error_Y(i), coefficient of error_X(i) and error_Y(i)
要做least square linear regression,assume error is norm distribution
怎么计算probability of fit? |
|
m*****e
发帖数: 268 | 45 很差吗？我看一些大牛教授也在上面发文章的啊
还有2本怎么样呢？
Journal of Applied Probability
Advances in Applied Probability |
|
w******w
发帖数: 92 | 46 A bug start from node A of Tetrahedron denoted by ABCD. It has equal
probability to move to other 3 nodes at any node. Ask what is the
probability of it has made 7 moves after returning to node A.
My answer is 164/729. 不知道对不对。 |
|
b***k
发帖数: 2673 | 47 ☆─────────────────────────────────────☆
leephy (leephy) 于 (Mon Feb 18 21:12:43 2008) 提到:
1) suppose U is a continuous uniform [0,1] random variable. what is the
probability that the decimal expansion of U contains no fives?
2) a stock is currently worth $100, assume that each of the next 10 days the
stock either increases or decreases in value by $1. what is the probability
that over the next 10 days, the stock reaches a maximum value of $104 and
on day 10 it sells for $100?
☆────────────── |
|
p********0
发帖数: 186 | 48 If we assume a flat term structure/risk free interest rate is known R,
assume we also two possiblity of the price changes, u and d
the Risk Neutral probablity = (R-d)/(u-d). We can back out using binomial
tree.
But if we donot have the flat term structure assumption,
should we always assume the Risk Neutral probability to be 1/2? |
|
y*****d
发帖数: 415 | 49 You already lost 12 times on a slot machine that has a payout rate of 96
cents on the dollar and the probability to win on the first run is 1%. What
is the probability that you will win the 13th time? |
|
o****u
发帖数: 401 | 50 The probability of performance falls out of the 99% VAR for 2 days in a
month?
I think this way, binomial distribution with x=2, p=0.01, and n=22(business
days in 08/07).
So p(x=2)=22!/(2!*20!)*(0.01)^2*(0.99)^20=1.89%
But how could I interpret it? The probability of one day occur is 1%. How
come 2 days become 1.89%?
Thanks |
|
