Archive for January, 2012

Twitter Updates for 2012-01-31

View CommentsWritten on January 31st, 2012 by caustic
Categories: Uncategorized

Get our FREE Open Source Historical Database by answering the 2 WORLD'S FASTEST TRADER/QUANT QUESTIONS

Post to Twitter

First ever FREE Quant Webcast using Google+ Hangout! Time 9am on Fri Feb 3/2012

View CommentsWritten on January 31st, 2012 by caustic
Categories: Quant Opinion
Tags: , , , ,
First ever FREE Quant Webcast using Google+ Hangout! Time 9am on Fri Feb 3/2012 Time 9am on Fri Feb 3/2012 This will be the first ever FREE webcast for QuantLabs.net in a Google+ Hangout. It will just be a general networking event to see who attends and have a general chat. Google+ is required but you can be in Google+ QuantLabs.net circle. I will start a Hangout at the specified time so logon Google+ then. I hope it works! NOTE: Time is in Eastern Standard Time! Accept my Google+ profile to get access https://plus.google.com/104809776202409789283

Get our FREE Open Source Historical Database by answering the 2 WORLD'S FASTEST TRADER/QUANT QUESTIONS

Post to Twitter

Events

View CommentsWritten on January 31st, 2012 by caustic
Categories: Uncategorized
Tags:
Here are upcoming Quant and HFT Meetup and Online Events
  • No upcoming events
AECv1.0 Created By Eran Miller

Get our FREE Open Source Historical Database by answering the 2 WORLD'S FASTEST TRADER/QUANT QUESTIONS

Post to Twitter

New Facebook group for Quant HFT high frequency trading with Matlab C++ Java R Python C# Excel

View CommentsWritten on January 31st, 2012 by caustic
Categories: Quant Opinion
Tags: , , , , , , , , , , ,

New Facebook group for Quant HFT high frequency trading trading with Matlab C++ Java R Python C# Excel VBA

I just created a similar group to the Linked In group.

http://www.facebook.com/groups/379703865378393  

Get our FREE Open Source Historical Database by answering the 2 WORLD'S FASTEST TRADER/QUANT QUESTIONS

Post to Twitter

Youtube video on how to learn quant models and algos FAST for hft through Matlab

View CommentsWritten on January 31st, 2012 by caustic
Categories: HFT High Frequency Trading, Matlab
Tags: , , , , , , , ,
Youtube video on how to learn quant models and algos FAST  for hft through Matlab I show why I use Matlab to quickly learn models and algos. This helps me quickly reverse engineer quality Matlab functions that are properly commented/documented from Mathworks. This is instead of pulling your hair out with poorly undocumented R packages or hand building your own C++ models.   I teach more by getting access here. YouTube Preview Image

Get our FREE Open Source Historical Database by answering the 2 WORLD'S FASTEST TRADER/QUANT QUESTIONS

Post to Twitter

Will SEFs Increase Market Fragmentation?

View CommentsWritten on January 31st, 2012 by caustic
Categories: Quant Opinion
Tags: , ,

Will SEFs Increase Market Fragmentation? Some firms see opportunities in electronic trading, while others express concerns about liquidity fragmenting across swap execution facilities, according to executives at TabbForum event. Fixed income executives at the likes of Goldman Sachs, BlackRock and Getco, among others weighed in on the outlook for electronic trading and central clearing at yesterday's TabbForum conference. But they also debated whether liquidity would increase or become fragmented among the various swap execution facilities (SEFs).

wallstreetandtech.com

Some firms see opportunities in electronic trading, while others express concerns about liquidity fragmenting across swap execution facilities.   == For more information, please visit our subgroup - The Dodd-Frank Act, Financial Regulations: Impact On Derivatives Technology Infrastructure (SUBGROUP) http://www.linkedin.com/groups?gid=4194682&trk=myg_ugrp_ovr    

Get our FREE Open Source Historical Database by answering the 2 WORLD'S FASTEST TRADER/QUANT QUESTIONS

Post to Twitter

The Rise of Cloud Computing for Quant Development on Wall Street

View CommentsWritten on January 31st, 2012 by caustic
Categories: Quant Development
Tags: , , ,
The Rise of Cloud Computing for Quant Development on Wall Street Looking to cut their capital expenditures on servers and applications, financial services firms are warming up to private clouds run outside their firewalls. As Wall Street continues to struggle with volatile markets, uncertain global economic conditions and vanishing profits, many firms are looking to reduce their capital expenditures. Targeting the costs of building data centers and maintaining server farms, more and more Wall Street organizations are looking to outsource pieces of their infrastructures to the cloud. While all of the largest financial firms already are experimenting with cloud technology in non-production areas such as server provisioning and storage networks, most remain cautious about security and refuse to let client data leave the relative safety of their own facilities. After facing sluggish profits and lower trading volumes last year, however, Wall Street firms are looking to convert large, up-front "cap-ex," or capital expenditures, into more variable "op-ex," or operational expenditures. As financial firms react to tighter IT budgets and the pressure to allocate resources to developing new products and entering new markets, the adoption of third-party cloud-based applications and services is expected to gain momentum in 2012.

wallstreetandtech.com

Looking to cut their capital expenditures on servers and applications, financial services firms are warming up to private clouds run outside their firewalls.   == For more info, please visit our subgroup - Financial Services Cloud Computing http://www.linkedin.com/groups?about=&gid=4167434&trk=anet_ug_grppro    

Get our FREE Open Source Historical Database by answering the 2 WORLD'S FASTEST TRADER/QUANT QUESTIONS

Post to Twitter

Return Forecasting via (Robust) Regression in quant analytics

View CommentsWritten on January 31st, 2012 by caustic
Categories: Quant Analytics
Tags: , , ,
Return Forecasting via (Robust) Regression in quant analytics High! Can anybody suggest a good reference paper on Return Forecasting via factor regression. In particular I am interested in the optimal choice of the return frequency (daily, weekly, monthly...) and on the pros and cons of using overlapping returns (say monthly or quartely) returns. In fact, I am interested in low turnover portfolios. Another question, does anybody knows any reference on robust regression methods applied to finance and return forecasting?     -- I'd suggest: Portfolio Selection with Robust Estimation http://www.est.uc3m.es/fjnm/esp/papers/RobustPortfolios.pdf It applies robust regression methods to compute optimal portfolios. Although nothing to do with return forecasting.   =- Daily returns understate beta. Monthly is better, if available. http://people.stern.nyu.edu/adamodar/New_Home_Page/AppldCF/derivn/ch4deriv.html   -- Chapter 6 of the book "Introduction to Modern Portfolio Optimization with NuOPT, S-PLUS and S+Bayes" by Bernd Scherer and Doug Martin talks about robust statistical methods and portfolio construction. Are you looking for something more specific?    

Get our FREE Open Source Historical Database by answering the 2 WORLD'S FASTEST TRADER/QUANT QUESTIONS

Post to Twitter

I have a very large NAS (IBRIX) environment and would like to find out how to locate who is using up all of the space

View CommentsWritten on January 31st, 2012 by caustic
Categories: Quant Development
Tags: , , ,
Question - I have a very large NAS (IBRIX) environment and would like to find out how to locate who is using up all of the space? Linux based commands are there however what tools are out there?   == If you have quotas enabled, you could do something like this in bash: for i in (LIST OF ALL USERNAMES); do echo $i; quota $i | grep FILESYSTEMNAME; done And then do some post processing. If you don't have quotas, then you're going to have to spider the filesystem in some fashion. Also, maybe IBRIX support (That's HP now, right?), might have something to recommend.   == EMC has some tools - VisualSRM, and NetApp has something to in the OnCommand. HP may have something, not sure. Try "agedu" also.   == my experience is that "spidering" (with du or otherwise) large remotely mounted filesystems (NFS or otherwise) is anything but cheap. quotas (user, tree, group , in some cases user under tree) in report mode (enforcement mode may be off) remain the easiest way under Linux, repquota will usually report for all users some filesystems (i am unfortunately not familiar with IBRIX) have specialized quota reporting commands    

Get our FREE Open Source Historical Database by answering the 2 WORLD'S FASTEST TRADER/QUANT QUESTIONS

Post to Twitter

Flash Solid state storage for HFT high frequency trading

View CommentsWritten on January 31st, 2012 by caustic
Categories: HFT High Frequency Trading
Tags: , , , ,
Flash Solid state storage for HFT high frequency trading Anybody have ideas how flash product such as Solid state storage could be applied to High performance computing? We are looking forward to work with HPC companies, is there a way to find out? Thanks.   == I have been looking into how to accelerate disk system performance with tiering of SSD and various speeds of spinning disk. I see it as a staging area for metadata or for frequently used small files. What are you looking for?   == We are looking forward to provide Solid state storage solution to all the HPC companies, do you think we are match?   == Please specify a product and a use case. I'll take a look.   == Standard 2.5"/SSD and SATADOM, our SATADOM has been qualified by Intel and Supermicro motherboard, it mainly used to boot up purpose. Our company web site, www.Innodisk.com   == From what I know quantum chemistry apps like to have not only large memory but also large and relatively fast swap on the nodes. This could be a small specialized niche. Example is a PC GAMESS (http://en.wikipedia.org/wiki/PC_GAMESS) . But in general I think you need to target SAN storage solutions.   == There are three tiers above SSD performance also, for those who can handle the price tag. Modest Performance Boost = OCZ RevoDrive http://www.newegg.com/Product/Product.aspx?Item=N82E16820227742 Even Higher Performance = Fusion ioDrive http://www.fusionio.com/platforms/iodrive-octal/ Unbelievable Performance = RAMSan-720 http://www.ramsan.com Each of these have their pros/cons. I believe the Fusion ioDrive cannot be booted with an OS, for example, but if you need fast, large storage, it bridges the price gap to the next jump up to the RAM San tier.   == What do you think about Tiering? Say front ending a large 1PB spinning disk installation by 10 or 20 TB of SSD? Then managing hotspots into the SSD?    

Get our FREE Open Source Historical Database by answering the 2 WORLD'S FASTEST TRADER/QUANT QUESTIONS

Post to Twitter

« Older Entries

Search Stuff

Subscribe to my Feed

Quantlabs.net

↑ Grab this Headline Animator

Follow me on Google+

Couldn't get data from google+

Upcoming Events

  • No upcoming events

Poll

Like Us

Blogroll

Latest Analysis

Latest Analysis

↑ Grab this Headline Animator

Ad

Sidebar Ad

Follow

Get every new post delivered to your Inbox

Join other followers: