m**k
发帖数: 18660 | 1 【 以下文字转载自 Fitness 讨论区 】
发信人: baton (rouge), 信区: Fitness
标 题: I think knee damage from jogging is over estimated
发信站: BBS 未名空间站 (Tue Jun 12 10:04:21 2007)
There's no question that exercise, especially running hurt knees. However,
most people who are suffering from the injury now are in their 50s. What
has been changed dramatically in the past 20 years are running shoes design
philosophy. Nowadays running shoes are cushioned for over-weighted people.
Most runners are fit or slim, and most runn |
|
K****D
发帖数: 30533 | 2 boxofficeguru's weekend estimate: $43-46m. Still in the game to catch
Titanic. |
|
T********r
发帖数: 6210 | 3 真给新娃说中了，E-mail附件的form上打印了$450修理费加$40回程运费。看到form心
想，真tm黑呀，就软件不认扫描仪这么一个小问题要这么多，连回程运费都还要我自己
出。
登到网站上去approve，发现Estimate被改了，修理费$45，回程运费$12.45，不知道这
是不是最后的价格？ |
|
i*****t
发帖数: 2435 | 4 我也收到新的estimate, 4/25-6/4.
3／15的金盒。
D7000 |
|
h*h
发帖数: 845 | 5 我3/21 order的，本来是estimated 4/18-5/3，但是三天前update payment method后
变成email notice了，到现在也没有update回去。要是谁也是同一时段order的已经
deliver了的话，烦请通知一声。 |
|
k******8
发帖数: 913 | 6 我差不多是同样的情况。上个星期update了shipping method，本来是estimated 4/21
~ 5/10，现在变成email通知，而且也没有变回去，很是不爽。 |
|
b*********n
发帖数: 5846 | 7 D800就不要等了，搞个革命机加套头好好旅游吧，拍纪念照够了。现在任何商家给的
estimated arrival date都是不靠谱的，连发布当晚下的订单都没有任何商家能给出靠
谱的到货时间，别说现在订的了。 |
|
e****t
发帖数: 17914 | 8 简单的例子 比如
y=a0+a1*x1+a2*x2+e
如何用grid search 来分类 data based on a1和a2的 estimated value？（e.g. a1<1
, a2>1） |
|
d********f
发帖数: 43471 | 9 【 以下文字转载自 History 讨论区 】
发信人: romo (romo), 信区: History
标 题: Re: 中国各个朝代的人均GDP estimates
发信站: BBS 未名空间站 (Sat Sep 28 21:02:17 2013, 美东)
看你怎么衡量了。
如果算工业制成品， 古代没有工业化，肯定吃亏， 一把菜刀都要好几十快钱。
算农产品的话， 古代生活还不错， 明朝南方一个农民家庭光种粮食就有1w5的年入，
如果搞些经济作物还做小工还要高好呢多。 江南一带几乎家家都有织床的。 这样算人
均3000美刀都有可能， 人吃的还都是organic的。
算房地产的话， 武大郎没几年靠卖饼（或者是馒头）就可以攥一小楼了。 就算一个小
县城估计都有10多万了。 |
|
c******g
发帖数: 1611 | 10 【 以下文字转载自 Chinook 俱乐部 】
发信人: cunzhang (慢羊羊), 信区: Chinook
标 题: nexus one selling estimation
发信站: BBS 未名空间站 (Thu Jan 21 15:52:04 2010, 美东)
I see on ebay it is still selling with good price, I would gues it sold
about 400k so far. |
|
f*****e
发帖数: 5177 | 11 I run SQL query estimated execution plan again one query.
The plan showed that the cost of "delete" and "insert" is 900%.
What does that mean?
PS: Which one of the following queries is better?
Query1:
if exist (select * from table1 where ...)
begin
update table1 set ...
end
else
begin
insert into table1 ...
end
Query2:
if exist (select * from table1 where ...)
begin
delete from table1 where ...
end
insert into table1 ...
Query3:
delete from table1 where ...
insert into table1 ... |
|
x*****a
发帖数: 48 | 12 早就说听说Lenovo家不靠谱了 想不到这么夸张
place order之前的Estimated Ship Date 还是6/24呢
我第一次下单买Lenovo的本子 有没有必要cancel掉啊 等着新本子急用啊 |
|
s********i
发帖数: 17328 | 13 好久没买笔记本了，问问Lenovo laptop的ship date estimate靠谱么？ |
|
a***m
发帖数: 74 | 14 I am confused by the results --
I tried the GARCH toolbox in Matlab. Compared to the results using
fminsearch, which I provided the likelihood function, the estimated
parameters are close for some cases, but very different for others, although
they both seem in reasonable ranges.
Could the differences be caused by the search methods themselves? |
|
x******i
发帖数: 148 | 15 Hello, everyone
Can anyone give an estimation cost for a new development stage company
opening balance sheet auditing. no any transacition, cash for investment
only. How much will cost?
Thanks a lot for the auditing experts.
Xbeartai |
|
g*********3
发帖数: 177 | 16 Hi All,
Recently I am doing CHIP-QPCR to test one TF binding. I use 10% of sample as
Input and the rest 90% to do IP. I have the TF-positive cell line A and TF-
depleted cell line B.
Primer is flanking the targeting site of TF.
The result I have is:
In A: Ct in IP 26 & Ct in Input 20 [Ct difference is 4]
In B: Ct in IP 30 & Ct in Input 20. [Ct difference is 10]
It seems TF binds to the targeting site.
However, I followed the calculation process of other people to estimate
yield:
1/9*1/(2^6)
whic... 阅读全帖 |
|
p*******t
发帖数: 501 | 17 【 以下文字转载自 Economics 讨论区 】
发信人: prescient (星辰大海), 信区: Economics
标 题: 问个maximum of simulated estimation问题
发信站: BBS 未名空间站 (Fri Mar 19 20:31:21 2010, 美东)
当在生成好simulator，计算出了对于每个choice的对应的likelihood function的值以
后，下一步就是去找parameter去maximize这个likelihood function的值，无论是用qu
asi-newton还是newton raphson，都要知道函数的first order或者hessian matrix的值
。在这种情况下，应该怎么去找这些的值呢？是不是简单的用倒数定义计算一下在那个
点的各个dimension的斜率就行了呢？
更规范的做法是什么呢？ |
|
z***a
发帖数: 427 | 18 我想问下如果estimator预算是1M，最后需要3M的话这个责任到底谁付啊？ |
|
|
c********r
发帖数: 73 | 20 Consider a sequence X1,X2,......,Xn of Bernoulli trials, Here the Xi are iid
as X with Prob(X=1)=p,0
show that there exists no unbiased estimator for the odds ratio p/(1-p)
due next Thesday.
thank jiay |
|
x***e
发帖数: 62 | 21 proof by contradiction;
suppose you can construct a function:
F(x1,x2....xn) as your estimator
calculate the expected value of your F(x1,x2,...xn);
E(F(x1,x2,...xn)); show that can never be p/(1-p);
it is getting nasty;
E(F((x1,x2...xn))=F(0,0,..0)*(1-p)^n+sum(F(Xi=1))*p*(1-p)^(n-1)+.....
+F(1,1,...1)*p^n;
and since the only possible value for Xi's are 0 and 1;
so F(0...0)=0; F(Xi=1)=1 or O; F(Xi=1,Xj=1)=0,1,2; .....
F(1,1,...1)=0,1,...(n/2)^(n/2);(if n is even)
and then you just calulate those |
|
J**Y
发帖数: 34 | 22 We can have more compact way to think about this. Suppose f(X) is an
unbiased estimator of odds ratio, where X=(x1,....,xn). Then
E[f(X)]=Sum_k[f(k)*C_nk*p^k*(1-p)^n-k], where C_nk is the combination
symbol. Because f(k)*C_nk is constant, let a_k denote it. Then
E[f(X)]=Sum_k[a_k*p^k*(1-p)^n-k]=p/1-p =>
Sum_k[a_k*p^k*(1-p)^n-k+1]=p, you then can show the equation
doesn't exist. |
|
e******e
发帖数: 274 | 23 I am using a Bayesian method to estimate a multinomial probit model through
data augmentation. All the priors are conjugate priors. Covariance matrix
follows wishart. Normalize the first element of covariance matrix to be 1. I
find that other elements of the covariance matrix are very easy to explode.
They are very sensitive to the starting value and the parameter of the priors.
Can any Da Xia tell me why those elements explode and how to deal with them?
Thank you very much. |
|
J**Y
发帖数: 34 | 24 My experience is that the values of latent variable from truncated normal
are easily to explode. You may want to add some truncations into your code.
Additionally, since you normalize the 1st diagnol element of covariance matrix
as 1, you can not still assume this matrix is wishart distributed. You can
check a recent paper (2000) in J. of Econometrics to see how to estimate
this kind of identified MNP model by Bayesian.
An easy way is to work on unidentified model, in which you specify priors
on |
|
c***9
发帖数: 16 | 25 I have a SUR model with panel data sets.
With trouble to find a suitable estimation software/package.
DataEstimate Panel Data: what software do you use? What packages? thanks a
lot. |
|
l******n
发帖数: 1204 | 26 请推荐有关bayesian estimation of DSGE models的讲义，教材。
希望有具体的讲解,详细的算法。
看了Fabio Canova的讲义还是不清楚。谢谢。 |
|
j*****e
发帖数: 333 | 27 正在看一篇文章，被分文章时editor也没有提这个，只有write up的文章。所以来咨询
一下。文章是estimate一个复杂的模型，对方有比较详细的解释和图表，我该要求作者
发给我他们的程序，我自己尝试跑一下么？谢谢建议 |
|
f***5
发帖数: 1569 | 28 奇怪，你referee别的类型的文章也会找人要code，或者数据自己重新estimate一遍么？
你觉得别人可能给你么？
这是你第一次写referee report么？
是的话建议你google写referee report都要做什么。 |
|
a***n
发帖数: 578 | 29 Is there a way to estimate wheter a polymer is expensive or not? For example,
given two polymers, how to judge which one is cheaper? Thanks in advance |
|
c********r
发帖数: 154 | 30 【 以下文字转载自 Chemistry 讨论区 】
发信人: colloidpar (macro), 信区: Chemistry
标 题: how to estimate experimental error?
发信站: BBS 未名空间站 (Wed Apr 9 09:54:59 2008), 转信
I saw many papers indicated the uncertainty in graphs. From my opinion, you
have to repeat the experiment several times and know the statistical error.
However, I don't think people have done that for every experiment. So how
did they add the error bar in the graph?
Thanks a lot! |
|
g*****g
发帖数: 45 | 31 If I have obtained some experiemtal data y_i(x), i=1, 2,...n from different
trial.and have calculated the average value and the standard deviation,
say y(x)=y_avg(x)+/-SD(x).
Now Iconvert y_avg(x) from linear domain to logarithmic domain,
y_avg_log(x)=log[y_avg(x)], how should I estimate my SD in logarithmic
domain?
May I use error propagation formular?
Thanks! |
|
c*******h
发帖数: 1096 | 32 【 以下文字转载自 Statistics 讨论区 】
发信人: cockroach (冬冬), 信区: Statistics
标 题: maximum likelihood estimation
发信站: BBS 未名空间站 (Fri Jul 29 11:15:29 2011, 美东)
怎么知道MLE的结果是可信的呢？
我手上有一堆数，假设是iid正态分布出来的。那么我可以用MLE
估计出分布的均值和方差。好，结果我很满意。完事，收工。
但是我怎么确定我的假设是对的呢？万一我手上的那堆数不是正
态分布出来的呢？无论怎么样MLE都可以给我算出均值和方差。
怎么检验我的关于分布的假设是对的呢？ |
|
|
G**********n
发帖数: 973 | 34 Bristol Names Cornelius Chief, Beats Profit Estimates (Update2)
By Lisa Rapaport and Michelle Fay Cortez
April 26 (Bloomberg) -- Bristol-Myers Squibb Co. named interim Chief
Executive Officer James Cornelius to the job until 2009, fueling speculation
he is preparing to sell the drugmaker.
Cornelius, 63, replaced the ousted Peter Dolan in September and until today
said he didn't want the job permanently. Last year he engineered the $27.5
billion sale of Guidant Corp. to Boston Scientific Corp.
`` |
|
p********0
发帖数: 186 | 35 Hi,
For GARCH model
\sigma_{t} = a + b * \sigma_{t-1} + c* \return_{t}, Can I run a linear
regression
to get the parameter? Can I use the regular linear regression since the \
sigma_[t-1} is
a response variable also a X variable? CAN I USE THE REGULAR regression?
Or to use a Maximum likelyhood function/get a derivative first, plug in the
\sigma_{t-1}
to \sigma_[t}. In this way I can get a parameter estimate? |
|
g****3
发帖数: 49 | 36 GMARCH? confused me for a while.
first of all, you can not use traditional or regular OLS regression, since
GARCH is similar with ARMA model, where h(t)^2 and h(t-1) are autoregressive
, also ht^2 and r(t-1)^2 are like moving average model. There are internal
dynamics.
Most people use MLE for parameter estimate, and Hull use it in his book,too.
second, I don't understand what you talk about in your Maximuim likelihood.
can you say it in human language rather than computer language.
third, it see |
|
p********0
发帖数: 186 | 37 For the state-space linear dynamic system,
y_t+1 = A_t * x_t + R
x_t+1 = B_t * x_t + Q
I read some paper about using EM - maximum likelihood to estimate the
parameter A and B
In simple case assume A_t = A_t-1 = ... A_1, we can maximize \hat{x_t} = E[x
_t | Y_1,,,n]
How do we handle the situaion A_t change over time, like ARMA(1, 1) case,
just use fewer observation Y_t-1, Y_t to estimat the A_t/B_t ??? |
|
p********0
发帖数: 186 | 38 Ding,
I may not be clear about my question, let me rephrase it.
For time series data, state space model, if the parameter is not static,
change over time
X_t = F_t * X_t-1 + R.
Y_t = H_t * X_t
What's the best way for finding the online parameter?
I found resurve Least Sqaure Error/Gaussian Newton method/Kalman filter all
can be used to do the online parameter estimate, what's the best method?
Anyone has experience?
[x |
|
l****o
发帖数: 2909 | 39 try shrinkage estimation. |
|
L*******t
发帖数: 2385 | 40 假如说我有一个options pricing model，很复杂。只能用MC实现。但是我不希望用
underlying历史数据来calibrate，有办法用option data来calibrate吗？
还有就是这个模型如果是一个结合参数和非参数的模型，有没有办法用期权数据来
estimate？
先谢谢各位大牛啦 |
|
C***7
发帖数: 241 | 41 Hello,
Iif I just know the electrical conductivity in an electrolyte,
how to estimate its ionic strength?
Many thanks! |
|
|
p********0
发帖数: 186 | 43 Hi,
I have the following question and wonder if anyone have some input?
INPUT
(X1, X2, X3,,,XN)
each data point X1 compose of 3 variables (v1, v2, v3) variables are
continusous like (1.5, 1.3, -0.3),
Then I have flag for (X1, X2, X3, ..., XN),
(1, 0, 1, ..0) to indicate a property of X,
assume Result (1 or 0) = (Av1 + Bv2 + Cv3) > threshold,
how do I estimate paramter A, B, and C?, Can I use E-M algorithm etc?
Thanks |
|
B****n
发帖数: 11290 | 44 你應該先問普通kernel density estimator的bandwidth和n的關係
然後bootstrap 的sample需要比較大一點的bandwidth 因為standard bootstrap
是取完放回再抽 因此要smooth的時候 需要比原來的sample大的window |
|
h***i
发帖数: 3844 | 45 那是不是说
standard bootstrap 和bootstrap from kernel density estimator with bandwidth
=n^{-1/2} 是一码事？
多谢，想破头了，实在是愚钝
这是for quantile variance
我翻了些paper, 比如 Angelis 和 Young的文章，没理出头绪 |
|
q********i
发帖数: 795 | 46 1. to calculate ors for each race, you dont need interaction, simply do
separate logit for each category using BY statement.
2. to get the or per 4 ng/nl, just multiple the point estimate by 4, and
then exponential it, same with the se |
|
h*********o
发帖数: 151 | 47 如题，ex，y = f(x,z)+e 怎么做local linear estimation啊？用什么package什么
function啊？谢谢啦！ |
|
c***9
发帖数: 16 | 48 Any body has used PLM in R?
Is it handy to estimate panel data? advantages and disadvantages.
If more think it is bad or awkward, then I won't spend time to look into it.
Thanks. |
|
p********0
发帖数: 186 | 49 Hi,
For time series data, state space model, if the parameter is not static,
change over time
X_t = F_t * X_t-1 + R.
Y_t = H_t * X_t
What's the way for finding the online parameter?
I found resurve Least Sqaure Error/Gaussian Newton method/Kalman filter all
can be used to do the online parameter estimate, what's the best method?
Anyone has experience? |
|
