Serving the Quantitative Finance Community

Search found 104 matches

  • 1
  • 3
  • 4
  • 5
  • 6
  • 7
by bostonquant
March 24th, 2008, 1:54 am
Forum: Technical Forum
Topic: Tools for woking with very large data sets
Replies: 8
Views: 58627

Tools for woking with very large data sets

<t>QuoteOriginally posted by: KludgeDudeHello,I'm currently working on a equity security (stocks) return archive. This archive will contains returns over multiple periods and currencies while using various calculation methods and data from various vendors. The only SQL tools I have available are lim...
by bostonquant
March 21st, 2008, 2:36 am
Forum: Trading Forum
Topic: How do you know if you are overfitting?
Replies: 2
Views: 58666

How do you know if you are overfitting?

Look at MSE (Mean Squared Error). The MSE of the overfitted model will be higher than the true model.
by bostonquant
March 13th, 2008, 3:10 am
Forum: Student Forum
Topic: career or wife
Replies: 48
Views: 64613

career or wife

Post was a bit harsh...
by bostonquant
February 25th, 2008, 5:34 pm
Forum: Student Forum
Topic: An interview question
Replies: 5
Views: 58880

An interview question

<t>QuoteOriginally posted by: circumflya1,a2,a3..... a_n satisfy the following conditions:1) a_n=1+1/a_(n-1), for n=2,3,4....n;2) a_1=1/a_n.What can you say about this sequence?This question is exactly what is wrong with today's quant interview. I can't tell you how many times I see a question like ...
by bostonquant
February 22nd, 2008, 6:50 pm
Forum: Student Forum
Topic: Normal Distribtuion Fundamental Question
Replies: 7
Views: 58939

Normal Distribtuion Fundamental Question

<t>QuoteOriginally posted by: AtreidesIn three words: Central Limit TheoremIn one word ... INCORRECTIt is not always obvious which of the standard distributions should be used when modeling data and a distribution with approximately the right shape (moments,skewness) should be looked for. For exampl...
by bostonquant
February 20th, 2008, 4:17 pm
Forum: Student Forum
Topic: probability
Replies: 1
Views: 58557

probability

rather than posting your homework questions out here why don't you a) interact with your fellow grad students b) go to office hours c)interact with the TA. This is a question from a 1st semester grad course in Probability. You're just depriving yourself.
by bostonquant
February 17th, 2008, 4:02 am
Forum: General Forum
Topic: Any Mac Users?
Replies: 15
Views: 62211

Any Mac Users?

<t>I saw the Macbook Air today. I just don't understand Macs. I own two Lenovo X61 ultraportable laptops Each weighs <3.1 lbs. 2.2 gHz Core Duo. 7200 rpm harddrive. 2 GB RAM. One I run linux the other Windows XP. The laptop running linux performance wise would destroy the Mac. Macs are completely ov...
by bostonquant
February 15th, 2008, 7:52 pm
Forum: Trading Forum
Topic: Barra
Replies: 15
Views: 70732

Barra

<r>QuoteOriginally posted by: eye51Hi, you can find it at:<URL url="http://www.alacra.com/alacra/help/barra_handbook_US.pdfIf"><LINK_TEXT text="http://www.alacra.com/alacra/help/barra ... k_US.pdfIf">http://www.alacra.com/alacra/help/barra_handbook_US.pdfIf</LINK_TEXT></URL> you are interested I can...
by bostonquant
February 15th, 2008, 7:35 pm
Forum: Technical Forum
Topic: good portfolio optimization and management approaches...
Replies: 7
Views: 59723

good portfolio optimization and management approaches...

<t>QuoteOriginally posted by: losemindHi all,It seems that no matter how researchers improve the estimation techniques for mean and for variance, most funds underperform the benchmark. Is add more exotic/structured products the only way to go to improve performance? Thanks!Concerning equities:1) Fee...
by bostonquant
February 15th, 2008, 12:08 am
Forum: General Forum
Topic: efficient frontier in excel
Replies: 5
Views: 61782

efficient frontier in excel

<t>I create efficient frontiers often using R. The function portfolio.optim() in the package tseries is useful.Consider you have a:Matrix of returns (in this example they are daily) called intlEqVector of holding constraints called max and minr<-NULLtemp<-NULLr<-seq(.00046,.00052,.000001)return<-NUL...
by bostonquant
February 13th, 2008, 3:53 pm
Forum: Student Forum
Topic: Simple matrix algebra question
Replies: 5
Views: 59491

Simple matrix algebra question

Find a vector z such that a . b' . z = cThis is quite simple
by bostonquant
February 9th, 2008, 11:03 pm
Forum: Student Forum
Topic: nb of variables in a regression
Replies: 3
Views: 59524

nb of variables in a regression

<t>QuoteOriginally posted by: JosephFrankHi,I have a small sample of 96 observations. I developed an empirical model with 4 independent variables to explain the independent variables and the results make sense. A referee asked me to add additional 9 variables to the model to test the robustness of t...
by bostonquant
February 9th, 2008, 11:02 pm
Forum: Student Forum
Topic: nb of variables in a regression
Replies: 3
Views: 59524

nb of variables in a regression

<t>QuoteOriginally posted by: JosephFrankHi,I have a small sample of 96 observations. I developed an empirical model with 4 independent variables to explain the independent variables and the results make sense. A referee asked me to add additional 9 variables to the model to test the robustness of t...
by bostonquant
November 6th, 2007, 8:01 pm
Forum: Careers Forum
Topic: Statistics Courses
Replies: 8
Views: 64309

Statistics Courses

If you aren't going to sit in the class why not buy a text and study whatever your interests are at your own pace? There is really no point in taking a course unless you want a degree and I am not aware of any on-line Stat programs..
by bostonquant
October 17th, 2006, 3:41 pm
Forum: Student Forum
Topic: Monte Carlo Simulation Option Valuation C++
Replies: 6
Views: 91157

Monte Carlo Simulation Option Valuation C++

Thanks everyone. Obviously this is academic, you wouldn't run this 50k times to get an estimate when you can very simply use B-S. I was asked to compare simulation vs. B-S. In any instance with the exception of speed it performs reasonably well.
  • 1
  • 3
  • 4
  • 5
  • 6
  • 7