Shane

less info
meta site favicon meta user | stack exchange network profile network profile
5,098 reputation
12444
bio website statalgo.com
location New York, NY
age 34
visits member for 2 years, 3 months
seen Apr 14 at 18:57
stats profile views 599

Quantitative researcher focusing on statistics and machine learning methods in finance. Primarily use R, C++, Python, various databases (including OneTick and KDB), and LaTeX on a daily basis.


116 Comments

suggestions reviews revisions comments badges posts accepts all
Apr
8
 comment FX Tick Data question
For reference: a quote is the top of the book bid/ask spread provided by a market maker or limit order book. In this case, all of Gain's data is quote data: they don't seem to provide transaction prices.
Apr
8
 comment FX Tick Data question
@Andy See my update; sorry, I was confused based on my prior experience with FX data. Don't use Gain myself, I use major banks and ECN's; Gain tends to serve retail clients more than institutions. See Mebane Faber's recent critique of Gain and company: mebanefaber.com/2011/04/03/currency-trading.
Apr
6
 comment penalizing negative skewness by linking $U(\mu)$ and $U(\Sigma)$
@Gortaur I just approved your edits. Thanks for improving it!
Mar
31
 comment better estimator of volatility for small samples
"As we all know, the best known sample estimator of volatility is the standard deviation of the log returns.". Really? Evidence?
Mar
20
 comment On my way to becoming a Quant
Voting to close. This kind of question has been closed in the past. This site is intended for professionals and researchers. See, for instance, meta.quant.stackexchange.com/questions/68/….
Mar
20
 comment Can the futures market's open interest predict commodity, treasury, and equity returns?
@ralph No need to close, just edit the title as @richardh suggests. As stated, open interest is some very specific.
Mar
20
 comment Can the futures market's open interest predict commodity, treasury, and equity returns?
I think that you misread the article. It says "open interest" (i.e. the number of contracts outstanding), which is totally different than the market opening.
Feb
19
 comment Technology stack used in Bloomberg
What reading? Any references would be helpful.
Feb
19
 comment Innovative ways of visualizing financial data
@Andy Neat! Maybe provide some of your experience as an answer?
Feb
19
 comment Innovative ways of visualizing financial data
@Andy For interactive web visualization, I find either (a) you use one of the commercial tools (which I believe can all read XML/JSON data) or (b) expect to put in a fair amount of effort. Another option is the google visualization toolkit mentioned below. Using either Protovis or Processing can result in amazing stuff, but they are both very low level, so expect to get your hands dirty. ggplot isn't really intended for that kind of usage (IMO). One of the vendors might provide it to you for free with some advertising (try Tableau first: tableausoftware.com/products/digital).
Feb
19
 comment Innovative ways of visualizing financial data
@Andy Which are you talking about specifically? Protovis is a javascript library, so that's straight forward. Processing is Java so that can be used from the web in any number of different ways. R doesn't have a specific web framework yet, but there are solutions (such as calling it from a foreign language interface). The commercial tools all have server versions.
Feb
17
 comment Which CEP platform is most popular for trading systems?
CEP is certainly on-topic, but I agree with @chrisaycock: this question isn't appropriate for the site. There is no way to answer which is the most popular, especially when you start to include things like open source solutions.
Feb
15
 comment What are some examples of non-financial risks and contingency plans?
To be clear, I never said this would be off-topic. I'm simply trying to narrow the scope of the question as it is written, or possibly get further information about what kinds of risks are under consideration. That being said, if someone else comes along and answers this: that's great.
Feb
15
 comment What are some examples of non-financial risks and contingency plans?
@Joshua I don't see how you tightened it up at all. You're still asking about any and every non-financial risk.
Feb
15
 comment What are some examples of non-financial risks and contingency plans?
This is a very broad question, including anything under the sun; can you tighten up your definition of non-financial risk a little?
Feb
14
 comment What are some computational bottlenecks that quants face?
I agree. Could be an interesting question if it was a little more focused. You tagged it gpu...are you asking what kind of quant problems can be improved by using a gpu?
Feb
14
 comment How to show that this weak scheme is a cubature scheme?
This is certainly on-topic here; it may just be that we don't have enough of an appropriate user-base yet.
Feb
11
 comment Free data on swap options
This is true for swap data generally; it's hard to get (hence the OTC part of it).
Feb
11
 comment Statistical learning libraries
@Zarbouzou Very wise decision.
Feb
11
 comment Statistical learning libraries
It is, generally, but can be very fast if used appropriately (vectorized functions, etc.). That being said: I always advise people to have a clear conception of how fast is fast enough for your given project before starting to make decisions. You will have a significantly harder time reproducing ESL in C++, so decide where you want to spend your time. So one common strategy is to do your research in R and then production implementation in C++ (when speed really matters).
1 2 3 4 5 6