|
n********y
发帖数: 48 | 2 我是个弱人， 虽然有PH.D，但是智力题啊，脑际急转弯什么都不行。 现在在一家小的
hedge fund做high frequency algorithm trading support engineer, 主要就是用
perl, vba写点工具，设计一些实验帮助提高系统latency, 也参与了些系统开发， 有
时候trader有事的时候能帮他cover一段时间。
我要求不高，也挺喜欢这样的工作。现在想换一个稍微大点的公司， 能够多见识一下
， 不知道大牛们的公司有没有这样的机会，能够帮助推荐一下。 |
|
x******l
发帖数: 104 | 3 Can anyone advice me some books?
I need Algorithms and programs (in Matlab, VB, VC++) for stochastic calculus
, multivariate calculus, linear algebra, differential equations, probability
theory and statistical inference.
I need some good books or resouces to understand how to program for finance. |
|
l****o
发帖数: 2909 | 4 top quant trader earn much more than top algorithmic trader. |
|
d*j
发帖数: 13780 | 5 quant trader <> algorithmic trader?
thanks |
|
l****o
发帖数: 2909 | 6 here quant trader normally means quant driven prop trader/arbitrager, but algorithmic trader do
electronic market making. |
|
l***m
发帖数: 920 | 7 false. algo trader do trade with automated program, but they don't
necessarily make markets. in most cases, the terms quant trader, algo trader
, systematic trader etc. are interchangeable.
algorithmic trader do |
|
l****o
发帖数: 2909 | 8 You are algorithmic trader? I am not but I go out with them a lot.
trader |
|
l***m
发帖数: 920 | 9 Said not every alto trader makes market. And not every high freq trader
makes market. In some markets, you make mor money by taking market.
actually write software everyday to do market making. Their software need
very good understanding of market microstructure and their algorithms varies
from time to time to meet clients' requirements. |
|
l***m
发帖数: 920 | 10 Btw from http://en.wikipedia.org/wiki/Algorithmic_trading
Algorithmic trading may be used in any investment strategy, including market
making, inter-market spreading, arbitrage, or pure speculation (including
trend following).
I promise you I haven't participated in the edition of this page. |
|
h******d
发帖数: 81 | 11 一个关于算法交易的疑问， 现在VWAP和TWAP是已经存在的算法还是仅仅是准则？如果
是准则，是不是已经有很多算法可以实现这两个准则？有没有什么资料可以参考的？我
好像很难发现algorithm trading 方面的材料。 谢谢！ |
|
m****r
发帖数: 141 | 12 【 以下文字转载自 JobHunting 讨论区 】
发信人: mitcar (mitcar), 信区: JobHunting
标 题: an interview algorithm question about finding even occuring freq
发信站: BBS 未名空间站 (Wed Jan 18 13:46:12 2012, 美东)
Design a method findEvenFreq(int[] a), which returns the single integer
value in the array which occurs with even frequency. Other numbers occur odd
numbers , which may be 7, 11 or even larger. All integers will be positive.
What if the integers are continuous.
For example, [1, 2, 3, 1] should return 1.
What if the in... 阅读全帖 |
|
p***o
发帖数: 836 | 13 Need to hire one Algorithmic Trading Quant urgently. Fresh/Junior/Senior
fine. MFE/PhD fine. Email to m***[email protected] Thx.
[发表自未名空间手机版 - m.mitbbs.com] |
|
p***o
发帖数: 836 | 14 收到一些简历，回答一些问题
---- full time position, 不是intern
---- 最好立刻上班
---- 一家国际投行，在manhattan, algorithmic trading/electronic trading
---- sell side
过了一遍简历，明天把整理出来的给manager.
谢谢大家关注；good luck. |
|
o********n
发帖数: 100 | 15 请问如果有若干只股票，大部分是从05年开始， 但其中有一些是在08年左右才出现，
如果我要用05年开始的数据估计covariance,则需要将08年的数据用em algorithm进行
补全。
这方面以前没做过， 请问各位大拿是否处理过相似的问题，有推荐的文章吗？ 搜了
一下，发现讲em进行数据补全的挺多，但是我用的矩阵是（股票*时间）的， 所以需要
考虑补全的时候是否一只股票一只股票处理比较好，还是将整体一起处理比较好。
谢谢！ |
|
W********g
发帖数: 22 | 16 ECN和 market maker在 execution algorithm 有什么区别？ |
|
A*******s
发帖数: 3942 | 17 Wondering if there is any easy-to-implemented algorithm to get the i-th
stochastic root of a transition matrix. Any reference? |
|
r*n
发帖数: 5 | 18 you have the algorithm right in your hand. just do the opposite of what
you did. replace subtractions in base 10 by additions in base 8. |
|
d*z
发帖数: 150 | 19
the len of r and s are no more than 2n
so
First we can get x+y=r-s+1 in O(n)
The time to get u=(x+y)(x+y)-4r is O(nlog(n)).
and the time to find x-y=sqrt(u) should be O(
n*log(n)log(n)log(log(n)) ).
so I think the time is O( n*log(n)*log(n)*log(log(n)) ).
Of course it is less than O(n*n). That's in polynomial time.
But in the fact, the comment algorithm to multiply and divid
is both O(n*n).
As to the sqrt(), we can use
x(n)=(x(n-1)*x(n-1)+a)/(2*x(n-1))
to found the sqrt(a), we only need to iter |
|
p*******p
发帖数: 13670 | 20 for calculating the eigenvectors and eigenvalues for a matrix? I need to
calculate a lot of eigvenvectors and values and use them in my program. Is
there any standard fast algorithm I can get online? Thanks. |
|
r***r
发帖数: 123 | 21 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! |
|
h******a
发帖数: 4 | 22 在不同条件下 如何用EM Algorithm去估未知参数
1。Gaussian Mixture Model with common known variance
2。Gaussian Mixture Model with common unknown variance
3。Gaussian Mixture Model with different unknown variance |
|
t*****9
发帖数: 19 | 23 Hi, I think if you go to "R search" and you will find some packages that
deal with such kind of normal mixtures with EM algorithm. What you need to
do is to specify those situations for the functions provided.
Hope this helps. |
|
f*********8
发帖数: 165 | 24 我刚开始学习R，90%的程序就是用package里的function。怎末才算是develop
algorithms?
另外能不能推荐一本教你用R自己写function的书啊。多谢。 |
|
f*********8
发帖数: 165 | 25 谢谢，我一直没搞清develop algorithms这个概念，原来我算是做过，就是比较简单而
已。 |
|
q**j
发帖数: 10612 | 26 最近看了一本有关于Bayes Analysis的书。讲到关键处，居然给了一个讲如何sample p
(beta,y_star|y)的算法。这里假设 y = X*beta + epsilon，y_star是一个用来
truncation的parameter。据作者讲这样的sample会比直接sample p(beta|y）简单。但
是我看着很不过瘾。看完书以后对最重要的application没有直接的感性认识。请问有
没有高手给指点一本讲如何sample p(beta|y）的书？就是如何设计，证明一个
algorithm 来sample p(beta|y)的。多谢了。 |
|
q**j
发帖数: 10612 | 27 我想打听一个具体的algorithm。多谢指点。 |
|
A*******s
发帖数: 3942 | 28 How to modify boosting algorithm according to a specific loss function? Any
big bull can recommend some readings? Thanks in advance. |
|
a***r
发帖数: 420 | 29 弱弱地问一个关于EM algorithm的基础问题，
大家都知道：(theta 为新一轮估计值，theta’为上一轮估计值)
L(theta) = Q(theta, theta')- H(theta,theta')
where H(theta，theta')=E(log[f(x|y,theta)]|y,theta')
为啥H(theta,theta’)总是小于等于H(theta',theta') 呢？
书上简单说了下，by Jensen‘s Inequality，
但是愚鲁的我没有想明白。。。
望版上大虾指点！ |
|
N******n
发帖数: 3003 | 30
algorithm
好像不算numerical solution, but it can be an approximation to numerical
optimization. |
|
y***n
发帖数: 51 | 31 我觉得Genetic algorithm 是算numerical optimization的。很多书本都只介绍比较传
统的方法，就想你前面列出的newton raphson. |
|
m**h
发帖数: 618 | 32 【 以下文字转载自 Military 讨论区 】
【 原文由 moth 所发表 】
My fellow engineers:
why can't we make a leaky bucket algorithm for anoying
posts? Set a counter on everyone, if there are more than
20 complians post to a certain ID, it get banned. and
every 10 days the counter decrease one. kind like what
they do with our drivers license.
And democracy my ass, BBS is only a place where people
yak yak and exchage information. It is not even close
to real society. |
|
d******4
发帖数: 132 | 33 neural network的这个back propagation algorithm算法，面试的时候会问什么样的问
题呢？多谢了。 |
|
a****k
发帖数: 117 | 34 You try to derive it by yourself. Then you have a better understanding of
the algorithm. If you just memorize the conclusion and some answers, experts
can figure out that. |
|
T*****u
发帖数: 7103 | 35 如果不考虑时间，我脚着可以；但是generic algorithm会不会太慢，而且也不一定会
有可以接受的结果出来？大侠们可以分享分享一下吗。 |
|
m**o
发帖数: 5261 | 36 我现在教学生，譬如SVM， 会具体描述算法，但学生有这个困惑，我会用PACKAGE 就行
了，为什么要知道具体的东西？
我过去在工业届的确需要自己写和改进一些ALGORITHMS。但对于大部分的学生而言，的
确好象没有必要知道。 只要会写CODE 和用现成的PACKAGE 就行了。
请问我该怎么回答这个问题? |
|
l****g
发帖数: 761 | 37 知道 details of a ML algorithm 是必须的
因为 ML 不像一般的程序， 错了就崩溃了
ML 搞错了只会给你个不符合原来预期的结果， 如果你不懂原理， 你甚至都不知道该
有什么样的预期
更何况很多场所还是需要自己制造/发明新ML工具， 尤其是涉及到 distributed
learning
但是我很不同意楼上的言ML必懂SVM细节之类的
聪明人做什么问题自然会把需要的技能捡起来
我周围有些同事以前都是搞算法理论的phd，没学过ML, 现在搞 ML 照样也能提出很犀
利的见解 |
|
m**o
发帖数: 5261 | 38 我现在教学生，譬如SVM， 会具体描述算法，但学生有这个困惑，我会用PACKAGE 就行
了，为什么要知道具体的东西？
我过去在工业届的确需要自己写和改进一些ALGORITHMS。但对于大部分的学生而言，的
确好象没有必要知道。 只要会写CODE 和用现成的PACKAGE 就行了。
请问我该怎么回答这个问题? |
|
l****g
发帖数: 761 | 39 知道 details of a ML algorithm 是必须的
因为 ML 不像一般的程序， 错了就崩溃了
ML 搞错了只会给你个不符合原来预期的结果， 如果你不懂原理， 你甚至都不知道该
有什么样的预期
更何况很多场所还是需要自己制造/发明新ML工具， 尤其是涉及到 distributed
learning
但是我很不同意楼上的言ML必懂SVM细节之类的
聪明人做什么问题自然会把需要的技能捡起来
我周围有些同事以前都是搞算法理论的phd，没学过ML, 现在搞 ML 照样也能提出很犀
利的见解 |
|
n*****3
发帖数: 1584 | 40 【 以下文字转载自 JobHunting 讨论区 】
发信人: nacst23 (cnc), 信区: JobHunting
标 题: 问一道(大)数据 algorithm
发信站: BBS 未名空间站 (Sun Mar 22 00:11:01 2015, 美东)
请教大家一下：
两组人， POSITIVE 和 Negative ，
say
POSITIVE 100K ppl，
Negative 900K ppl.
基本的数据结构 是 人的 ID 和 length of stay（待了几天）。
ID length of stay(days)
ppl-0000001 8
ppl-0000002 10
...
目的是 sample Negative 组 出来 100K 人 ,
which one-to-one match the Positive 组 人
的 length of stay（待了几天），
这样 match 完, 两组人的 100K 个 length of stay（待了几天）
完全... 阅读全帖 |
|
n*****3
发帖数: 1584 | 41 the for loop will take a long of time to complete...
want to find out some good algorithms for this. |
|
s******e
发帖数: 114 | 42 still confused.
you are looking for algorithm which is better then O(n). if sorting is only
for match sample, it is bad. sorting is kind of O(nlogn), I could be wrong
, if R uses bucket sort in this case.
assuming data was sorted already for other purpose, merge join is fast.
otherwise, I would use hash join. |
|
n*****3
发帖数: 1584 | 43 good point;
if everything in c/c++, I bet there is better solution;
but here I just want to speed up a small, but very time consuming(big loop)
part ,
in R.
Sure I can use std::sort to sort them in the RCPP code instead of using R/
data.table/setkey.
BTW do you have better algorithm for this? my solution is just for 抛砖引玉.
only
wrong |
|
d*****r
发帖数: 39446 | 44 【此篇文章是由自动发信系统所张贴】
algorithmics 已经退出本俱乐部, 特此通知. |
|
d*****r
发帖数: 39446 | 45 【此篇文章是由自动发信系统所张贴】
algorithmics 已经退出本俱乐部, 特此通知. |
|
|
d*****r
发帖数: 39446 | 47 【此篇文章是由自动发信系统所张贴】
algorithmics 已经退出本俱乐部, 特此通知. |
|
d*****r
发帖数: 39446 | 48 【此篇文章是由自动发信系统所张贴】
algorithmics 已经退出本俱乐部, 特此通知. |
|
f**d
发帖数: 768 | 49 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 |
|
d**********o
发帖数: 1321 | 50 第一个项目report
这时偶刚到CSAC工作不久，与小A同学还不熟，我用的还是latex。随着贴的作业越来越
多，应该是用有共同爱好的小伙伴更亲密些。这次贴latex，下次才再org-mode。
\documentclass[b5paper,11pt, abstraction, titlepage]{scrartcl}
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage{CJKutf8}
\usepackage{multirow}
\usepackage{multicol}
\usepackage{listings}
\usepackage{geometry}
\geometry{b5paper}
\usepackage{graphicx,floatrow}
\usepackage{graphicx,subfigure}
\newsavebox{\abstractbox}
\renewenvironment{abstract}
{\begin{lrbox}{0}\begin{minipage}{\t... 阅读全帖 |
|
