d******n
发帖数: 42 | 1 what do you need the inverse for? |
|
t***s
发帖数: 4666 | 2 rule #1 in matrix computation, avoid calculating inverse explicitly at
all expense. use QR or SVD.
to
However
zero |
|
|
l******n
发帖数: 4 | 4 or the mean and variance of the inverse of a Guassian RV?
Thanks |
|
|
y****g
发帖数: 14 | 6 I have a question.
Can the inverse Laplace transform of
[sqrt(s^2+c*s+k)-sqrt(s^2+c*s)]^n
where n is integer and c, k are real. transform from s->t, s is complex
be found?
Thanks for your help! |
|
L******k
发帖数: 2945 | 7 Using levinson-durbin recursion, you can calculate inverse of toep matrix
recursively, which at step n+1 involving calculating the reflection coeff.
and a n+1 dimension vector.
is |
|
J*****n
发帖数: 4859 | 8 google "inverse problem" |
|
m***a
发帖数: 2 | 9 Another question. can you recommend a list of course works that I should
take if I want to study inverse problem more systematically.
Thanks
mlfma |
|
|
mw
发帖数: 525 | 11 hi, math gurus:
I read that it is usurally preferable to inverse the triangle matrix L and U
, rather than the original matrix itself.
but I cannot understand why it is like this
can anyone here give a clue ? thanks a lot ? |
|
mw
发帖数: 525 | 12 thanks for your suggestion
yes, I know (roughly) how to calculate the matrix inverse manually.
so, do you mean that LU only improves the efficiency? why my impression is
that it is more robust to ill-conditioned matrix?
thanks |
|
c*******h
发帖数: 1096 | 13 in general no one would like to invert the L or U factors.
when you solve a matrix equation Ax=b, first LU factorize A, then do a
forward-solve on Ly=b followed by a back-solve Ux=y to obtain x.
The LU factorization is equivalent to Gauss elimination.
otherwise, how to compute a matrix inverse?
U |
|
r***r
发帖数: 123 | 14 BlackKnife,
Thanks for your expertise.
I know very little about statistics. Don't even know what is gamma distribution
. Would you please give me some formula or agrithom to calculate gamma
inverse. I couldn't find any.
We currently use Excel/SAS to get around of this. My boss want do it in the
Oracle.
Thanks once again!
自己
其他
新月 |
|
b*******g
发帖数: 513 | 15 最近学了点数论。知道了什么是multiplication table mod n.但是不知道inverse
table mod n.哪位大虾帮忙解释一下。谢了！ |
|
s*******t
发帖数: 2896 | 16 假设A是n by n实方阵，元素是a_ij。
Inverse(A)是一个映射，domain 是所有可逆方阵，这个映射的Jacobian是一个n^2 by
n^2的方阵，元素是a_ij的函数。
求这个Jacobian的determinent。好像结果还是det(A)，或者类似的东西。
老师说这个是well-known result，可是我不知道在哪本书上。
谢谢！ |
|
q**j
发帖数: 10612 | 17 比如 inverse(I+xy') = I - x'y/mu。我想告诉mathematica x 和 y，（只是符号，不
赋值)，然后让mathematica给我把 1-xy'/mu显示出来。请问这件事情如何去搞？多谢
了。 |
|
h**********c
发帖数: 4120 | 18 Hopefully 3*5 \triplequv 1 mod 2
3 and 5 are multiplicative inverse.
good luck |
|
j******n
发帖数: 271 | 19 I'm trying to find the inverse of the Vandermonde matrix (a_{ij}) with
a_{ij} = \omega^{(i-1)(j-1)}, 1<=i,j<=N
where \omega is a non-trivial root of z^N-1=0. Can someone shed some light
or point me to some references? |
|
p*****k
发帖数: 318 | 20 from wikipedia, reference 2:
Macon & Spitzbart, "Inverses of Vandermonde Matrices"
The American Mathematical Monthly, Vol. 65, No. 2 (Feb., 1958), pp. 95-100 |
|
o******l
发帖数: 35 | 21 How hard is it to calculate the inverse covariance matrix of multivariate
gaussians? Isn't it just basic linear algebra? |
|
l******n
发帖数: 9344 | 22 Hahaha...
How hard is it to calculate the inverse covariance matrix of multivariate
gaussians? Isn........
★ Sent from iPhone App: iReader Mitbbs 6.0 - iPhone Lite |
|
k*****y
发帖数: 744 | 23 正在看C++ design patterns and derivatives pricing。
有没有大牛能帮忙稍微解释一下Beasley-Springer/Moro的inverse normal算法，和后
面一个normal的cdf算法，那些常数大概是怎么得来的？Thanks. |
|
y**t
发帖数: 50 | 24 if a,b,c are very close,you can use
a^{-1}(1-ba^{-1}-ca^{-1}) to approximate
in fact,using algorithms from numerical ricipe,
it is not terrible to do inverse directly
unless your dimension is millions |
|
r***r
发帖数: 123 | 25 Please help me out of this.
I need a algorithm to write a function in ORACLE.
I know there is a function GammaInv in Excel and GamInv in SAS. Thoses
functions won't help me to implement it in Oracle.
I know very little about statistics. Don't even know what is gamma distribution
. Would you please give me some formula or agrithom to calculate gamma
inverse. I couldn't find any.
We are currently using Excel/SAS to get around of this. My boss want do it
in Oracle.
Thanks in advance! |
|
x*******i
发帖数: 1791 | 26 i remember there is transformation form to estimate two parameters of
inverse gaussian. i think you can do gibbs.
something like two roots of the function.... |
|
s***t
发帖数: 125 | 27 数学太差，实在不会，求助：
自己写一个很复杂的方程：
PPMM=function(f){...}，没法symbolic地求积分
那么想求其inverse fourier tranform的话，就是
inv_FT_PPMM=function(k)
{ integrate(PPMM*exp(2*pi*f*1i*k*f),-1,1) } (假设f的domain是[-1,1])
可是R为什么总报错呢，说non-conformable arguments，这是什么意思？
多谢！ |
|
h******3
发帖数: 190 | 28 You are still referring to inverse Wishart, right? |
|
g*******i
发帖数: 258 | 29 right,
some people do argue that inverse Wishart as a prior is not an ideal choice.
maybe you can try things like Jeffrey's prior, which I am not very sure |
|
A******s
发帖数: 427 | 30 SAS 和R都试过了。 算得inverse matrix和original matrix 相乘都不能得到identity
matrix. |
|
a******p
发帖数: 30 | 31 generalized inverse
identity |
|
l****z
发帖数: 29846 | 32 http://online.wsj.com/articles/shire-gets-revised-abbvie-takeov
AbbVie Nears $54 Billion Shire Deal
Deal Would Be Latest 'Inversion' Play
By Hester Plumridge
connect
Updated July 14, 2014 2:13 p.m. ET
LONDON—U.S. drug maker AbbVie Inc. ABBV -0.02% is close to clinching
a deal to buy Dublin-based Shire SHPG +2.23% PLC for more than £
31 billion ($53.6 billion), in what would be one of the largest so-called
inversion deals through which U.S. companies are seeking a lower corpo... 阅读全帖 |
|
c*******d
发帖数: 255 | 33 Munkres的"Topology(2e)"第21页第5题：
f: A->B 是一个映射，g:B->A称为left inverse if g f = id(A),
h:B->A称为right inverse if f h = id(B).
(a) If f has a left inverse, then f is injective;
if f has a right inverse, then f is surjective
(b) Give an example of a function that has a left inverse, but no right
inverse
(c) Give an example of a function that has a right inverse, but no left
inverse
(d) Can a function have more than one left inverse? More than one right
inverse?
(e) Show that if f has both a left inv |
|
S*****s
发帖数: 287 | 34 Inverse agonist?
http://en.wikipedia.org/wiki/Inverse_agonist
In pharmacology, an inverse agonist is an agent that binds to the same
receptor binding-site as an agonist for that receptor and reverses
constitutive activity of receptors.[1] Inverse agonists exert the opposite
pharmacological effect of a receptor agonist. Inverse agonists are effective
against certain types of receptors (e.g., certain histamine receptors and
GABA receptors[2]) that have intrinsic activity without the action of a
l... 阅读全帖 |
|
l*****r
发帖数: 2123 | 35 The inverted yield curve has set off alarm bells that a recession may loom
around the corner. Historically, that's been a good bet, and recessions
almost always unleash bear markets for the Dow Jones, S&P 500 and Nasdaq.
But this yield-curve inversion, which may be short-lived, looks different.
While it raises some concern about the long-term economic outlook, the
recent inverted yield curve actually is welcome in the near term.
"We may actually have seen a rather unique occurrence: the Fed actu... 阅读全帖 |
|
f**d
发帖数: 768 | 36 这是一本计算神经科学的优秀著作，全文拷贝这里（图和公式缺），有兴趣的同学可以
阅读
如需要，我可以分享PDF文件（－－仅供个人学习，无商业用途）
From Computer to Brain
William W. Lytton
From Computer to Brain
Foundations of Computational Neuroscience
Springer
William W. Lytton, M.D.
Associate Professor, State University of New York, Downstato, Brooklyn, NY
Visiting Associate Professor, University of Wisconsin, Madison
Visiting Associate Professor, Polytechnic University, Brooklyn, NY
Staff Neurologist., Kings County Hospital, Brooklyn, NY
In From Computer to Brain: ... 阅读全帖 |
|
g*******y
发帖数: 1930 | 37 这个问题可以通过考虑# of inversions来证明，楼上说的是对的
假设把前A个数，跟后B个数，交换位置
交换前,A个数跟B个数之间形成的inversions数是inv, 非inv数是~inv,
那么有
inv + ~inv = A*B (e.q.1)
交换后，原来的非inversion，变成了inversion，
那么
增加(或减少)的inversion数目 = inv - ~ inv (e.q.2)
可以看出，这题里面，A,B只能是{0,3},{1，2}，A*B必然是偶数
那么两个方程e.q.1, e.q.2给出整数解的必要条件就是 “增加(或减少)的inversion数目”为偶数
PS，考虑inversion数目是与次序相关的问题里面常见的一个topic |
|
A*V
发帖数: 3528 | 38 【 以下文字转载自 Stock 讨论区 】
发信人: AUV (滞胀·衰胀·萧胀), 信区: Stock
标 题: 熊市基金
发信站: BBS 未名空间站 (Tue Feb 27 17:32:21 2007)
如果大家准备作空的话，不妨看一看Rydex和ProFunds的Inverse MF。特别是在IRA/
Roth IRA等不能作空股票的帐户，Inverse MF是作空的最佳选择。
http://rydexfunds.com/ourproducts/product_lineup.shtml
Inverse Dynamic Dow
Seeks to provide investment results that inversely correspond to 200% of the
daily performance of the Dow Jones Industrial Average
Inverse Dynamic OTC
Seeks to provide investment results that inversely correspond to 200% of the
d |
|
g*******y
发帖数: 1930 | 39 推广一下，把matrix变为数组
index = i*N+j
rotate某一行，证明inversion数的改变是偶数，这个很trival
rotate某一列，证明数组中inversion数的改变是偶数，就稍微复杂一些，但是仍然可
以证明同样的结论
关键就在于意识到，在某个局部，一段子数组A紧跟着子数组B，只考率cross A,B的
inversion的话，如果A，B至少有一个个数为偶数，那么：
inv + noninv = even number成立
并且，交换A,B后，所有原来的inversion变成非inversion，vice versa
这样，inversion数目的改变量，始终为偶数
matrix. |
|
t*******r
发帖数: 22634 | 40 这个解题思路是数学里的一个小 trick （当然也是重要 trick），就是如果问题比较
难解，但该问题的 inverse problem 比较容易解，那么先解该问题的 inverse
problem，然后把结果再 inverse 回来。。。
这题稍微麻烦点的地方是 inverse 需要用到集合逻辑（类似 venn diagram），而给集
合逻辑数数要用到排列组合，这步 inverse 俺到现在也想不出啥好办法，于是就暴力
求解了 inverse 这步。。。 |
|
m*****P
发帖数: 1331 | 41 总结了一下下文中提到的几个
做空的ETF
EPV 2倍杠杆 跟踪MSCI Europe Index 波动相对下面两个大
EUO DRR 也是两倍杠杆 两者趋势差不多 波动相对EPV小 因为跟踪的是currency
其他可以自己short的ETF 包括（文中提到的）
EUFN VGK 还有针对某个国家的 BUNL ITLY
还有楼下朋友nothintomind 提到的：
You could opt to short EWG (Germany), EWP (Spain), EWI (Italy), EWQ (
France)
==
做空欧洲的ETF：
http://www.indexuniverse.com/sections/blog/9911-shorting-europe
Related ETFs: EUFN / DRR / EUO / EPV / VGK / BUNL / ITLY
The European debt crisis is the financial story that refuses to go away, so
investors may as well figu... 阅读全帖 |
|
s**s
发帖数: 404 | 42 来自主题: EnglishChat版 - 问个弱问题 Please refer to wikipedia:
http://en.wikipedia.org/wiki/Inverse_element
"Every real number s has a additive inverse given by -x. Every nonzero real
number x has a multiplicative inverse give by 1/x. By contrast, zero has no
multiplicative inverse."
"A square matrix M with entries in a field K is invertible if and only if
its determinant is different from zero. If the determinant of M is zero, it
is impossible for it to have a one-sided inverse; therefore a eft inverse or
right inverse implies th |
|
|
t******l
发帖数: 10908 | 44 其实这个 inverse problem，好像也是 pre-algebra 和 post-algebra 娃娃们重新排
队的因子之一。
在 pre-algebra 阶段，换成解 inverse problem 似乎是一种尖端技巧。在这个阶段，
只有被认为非常 “聪明” 的娃才会玩那种。
而在 post-algebra 阶段，看看 inverse problem 是不是更容易解，基本是一种
routine。而在这个阶段，咋 inverse 都不出错，以及只要有 inverse 就不会漏掉的
，这种思维严谨的娃，至少在入门级的时候，比单纯的 “聪明” 似乎更有优势。 |
|
T*******p
发帖数: 524 | 45 Technically the market is reaching a point so called a "support". Unless
some real disasters happen (such as a flash crash), the market should be
slowly
bouncing back in next few weeks. Dump everything if the market could reach
203, which could be the last chance to exit market safely for the next 2
years. If the SPY has passed 203, don't feel regretted, buy inverse ETFs, or
just walk away from market.I can guarantee you can get your stocks back at
a much lower price next year. How deep the mark... 阅读全帖 |
|
y******n
发帖数: 67 | 46 Inversion coefficient is used to identify the transistor operation
region, weak inversion (subthreshold region), moderate inversion and strong
inversion. Therefore, three design parameters are channel length, inversion
coefficient and drain current. |
|
l*****7
发帖数: 8463 | 47 美国的爱国教育来了:
私人公司还爱不爱国了？
政府不批，私人公司就是个屁！
http://www.ft.com/cms/s/0/c65488bc-91db-11e5-bd82-c1fb87bef7af.
Pfizer seeks Washington support for $160bn Allergan deal
David Crow and James Fontanella-Khan in New York and Barney Jopson in
Washington
Share Author alerts PrintClip Commentspfizer take over©Dreamstime
Pfizer’s chief executive has called on Washington to support its $160bn
takeover of Dublin-based Allergan, as politicians including Hillary Clinton
lined up against the drugmaker’s attempt ... 阅读全帖 |
|
l****z
发帖数: 29846 | 48 Medical Merger Part of 'Tax Inversion' Wave
Medtronic's Deal to Buy Covidien for $42.9 Billion Will Help It Sidestep U.S
. Corporate Taxes
Medtronic Inc. MDT -1.10% 's agreement on Sunday to buy rival medical-
device maker Covidien COV +20.45% PLC for $42.9 billion is the latest
in a wave of recent moves designed—at least in part—to sidestep U.S.
corporate taxes.
Covidien's U.S. headquarters are in Mansfield, Mass., where many of its
executives are based. But officially it is domicile... 阅读全帖 |
|
m**k
发帖数: 1503 | 49 https://www.donaldjtrump.com/positions/tax-reform
似乎砍得太狠了点，虽然比起4年前的已经好点了。看下来有点捏把汗，cut那么多，就
靠堵itemized deduction和海外资产抽税来补。
这个其实数据都有，cut之后会缺多少，itemized deduction能出来多少，剩下的全部
都要靠海外资产，可是海外资产到底多少这个就没底了，就算知道，怎么抽税也是个很
大问题，这个事情后面一定会被人攻击的。
trump自己也显得没信心，被问到的时候只说经济增长了税收就会多起来，但是这个就
显得很wishful thinking了，如果税收不够，就只能减福利了，这个估计也在他的计划
里面只是没提。
Too few Americans are working, too many jobs have been shipped overseas, and
too many middle class families cannot make ends meet. This tax plan
directly meets these challeng... 阅读全帖 |
|
g*******y
发帖数: 1930 | 50 我举个例子，
最初的矩阵写成数组：
1 2 3 4 5 6 7 8 9
如果要rotate某一列(不失一般性，另向左rotate一位)，例如，第二列 2 5 8
那么，考虑inversion的改变，那么有以下四处局部发生变化
a： (3 4) 5 -> 5 (3 4)
b: (6 7) 8 -> 8 (6 7)
c: 2 (3 4 6 7) -> (3 4 6 7) 2
以上a b c是把2 5 8和剩下的数分开来考虑
d: 2 5 8 -> 5 8 2（这个已经证明过了，inversion改变始终为偶数，所以d可以忽略
不考虑了）
观察a b c情况，()里面的个数永远都是偶数，为什么，因为同列隔行的两个数之间，
始终都是间隔了2个数。
那么对于a b c中任意一种情况，都有inversion数+非inversion数=偶数成立
证毕 |
|
