Basically, to get started I need only U.S. stock market data. It'd better to have real-time streaming, but 15m delayed feed is alright as well. Thanks.
I've done this too, but to be honest its not worth much. If you need minute by minute data then you probably need it to be real time (this is 12 minutes behind) however for most technical strategies you can use end of day data. For this I find google's data to be the most accurate, and easiest to scrape. here is an example url:
The base is http://www.google.com/finance/historical?q=
then the symbol such as goog
then you need to specify a start date by adding the query string "startdate" to the url
along with the query string "enddate"
and finally you will always want to dump it as a csv: "output=csv"
a note about the dates, It should be formatted in such a way that it is [Month][Day][%2C+][Year] where Month is the first 3 letters of the month.
I use DTN's NxCore for US Level I real-time data (last sale + bid/ask updates). The feed is excellent, but costs $500 per month. Here is the link: http://www.dtnmarketaccess.com/
I want to thank everyone who helped me to get all the information I need on this. Though I found the solution which fits me most(xignite.com 59USD/month for real-time quotes), I wouldn't be able to do it without you. Thank you.
12 comments
[ 3.0 ms ] story [ 37.9 ms ] threadhttp://finance.yahoo.com/q/hp?s=GOOG
There's a CSV download option.
The base is http://www.google.com/finance/historical?q= then the symbol such as goog then you need to specify a start date by adding the query string "startdate" to the url along with the query string "enddate" and finally you will always want to dump it as a csv: "output=csv"
a note about the dates, It should be formatted in such a way that it is [Month][Day][%2C+][Year] where Month is the first 3 letters of the month.
all together it would look like this: http://www.google.com/finance/historicalq=goog&startdate...
I should note that has been some changes since the last time I've ran the code to scrape the data, but i don't think things have changed dramatically.
also if you choose Yahoo http://code.google.com/p/yahoo-finance-stock-data/
You can have a system that works very well with Yahoo data but not in the real world due to two reasons:
-data integrity
-survivorship bias
So to get good data you need to pay. It's imperative you have good data.
A couple sites: Futures Data (free): http://www.tradingblox.com/tradingblox/free-historical-data....
Futures Data (paid): http://www.pinnacledata.com/
Stock Data (paid, highly recommended): http://www.premiumdata.net/
Stock Data (paid): http://www.csidata.com/
You can get forex data from oanda, fxcm, metatrader, ninjatrader.
You can also backtest on tradestation.
Also, tick data and OHLC data can affect your style-- if you're swing trading you need OHLC but if you're doing more short term, you need tick data.
It's free, and decent quality.
The interface is RESTful, but the data is a funky binary format that's hard to parse without the documentation they make you sign an NDA to get.
http://www.tdameritrade.com/tradingtools/partnertools/api_de...