Reputation
9,972
Next tag badge:
74/100 score
24/20 answers
Badges
1 7 33
Impact
~298k people reached

  • 0 posts edited
  • 21 helpful flags
  • 254 votes cast
May
20
comment Is there any thing out there as a substitute for KDB?
I was not aware of that, thanks a lot for this update, +1
May
20
comment Databases for storing and querying high frequency tick-level data?
ups, just saw your answer, also recommended it in my comment to the question.
May
20
comment Databases for storing and querying high frequency tick-level data?
@Arman, check out Teafiles, also consider writing your own binary data store, or you could consider document DBs, but other than that there are not a whole lot of open-source columnar DBs out there.
May
19
comment Databases for storing and querying high frequency tick-level data?
out-of-the-box? This concept does not exist in finance, unless of course you want to "trade" bitcoins. Whatever time series store you decide to go with you will not get around doing a lot of performance analytics, tweaking, and customization. Just for your reference, KDB generally sends a host of "consultants" who stay for several days just to setup the initial database structure.
May
18
comment Black-Scholes: Why the focus on volatility?
@DonShanil, are you addressing the above to me or Andrew?
May
18
awarded  Custodian
May
18
reviewed Leave Closed Source of Quandl Open Data
May
17
comment Difference betweem martingale property and adapted filteration
just curious, does Steven Shreve ring a bell for you?
May
17
revised Looking for C# library that provides/contains performance analytics
added 54 characters in body
May
17
revised Looking for C# library that provides/contains performance analytics
added 5 characters in body
May
17
comment Looking for C# library that provides/contains performance analytics
Thank you very much for pointing me into the rredis direction and indirectly into how to distribute workloads using different R sessions. I have not used R for quite some time as explained in my own answer, but have to admit that especially the capability to distribute workload through multiple R sessions opens up a whole load of new opportunities. Yet, for this specific project I still decided to write my own customized solution (also explained in my own answer). But that in no way diminishes my appreciation for walking me and other users through the options available to connect with R.
May
17
accepted Looking for C# library that provides/contains performance analytics
May
17
answered Looking for C# library that provides/contains performance analytics
May
16
awarded  Benefactor
May
15
comment What is the equation for Garman-Klass volatility?
Google is your best friend, todaysgroep.nl/media/236846/measuring_historic_volatility.pdf
May
15
comment Looking for C# library that provides/contains performance analytics
that itself would not be the issue, I just wonder whether there is a simple way in R to execute "string commands". I think I could use Parse and Eval...?
May
15
comment Looking for C# library that provides/contains performance analytics
I see that, but I fail to see how I can perform automated command processing off redis queues with rredis. Is my assumption correct that I would have to write the listener and processor myself?
May
15
comment Looking for C# library that provides/contains performance analytics
I posted a question on stackoverflow (stackoverflow.com/questions/23672491/…) because I am getting errors when executing a simple example regarding doRedis. Is doRedis even the right package to use if I just want to have a single worker that picks up commands from the redis queue and stores the results back on redis? I am asking because doRedis seems to really zero in on distributed workload processing through foreach loops. Any hints? Thanks
May
15
comment What is the motivation for index benchmark?
And your last point, while valid, only applies again to relative performance. Yes you may get fired if you underperform all your peers, but nobody questions you nor fires you if even the top performer lost a sizable percentage of money invested. Because of that hardly anyone in the conventional buy side industry (hedge funds excluded) truly searches for outstanding but unconventional trading ideas. I do not only fault the PMs, it is the whole regulatory framework that is to be blamed. The term "Long only" should not even exist in our industry.
May
15
comment What is the motivation for index benchmark?
But are you not omitting the actual metric, absolute returns? The issue is that most portfolio managers and almost the entire buy side industry hides behind relative performance. Because of this the incentive to produce alpha diminishes because it would mean that you have to take on risk or take a specific non-conforming view. But fact is that regardless of performance, whether positive but most often underperforming indexes or absolutely negative, relative performance to indexes is touted as a means to market such funds.