Python real time data feed

[Lacanau]

So i've done a couple of courses on computational finance, and i'm pretty familiar with python so I want to start applying some of what i've learned on actual real time data feeds rather than downloading and computing historical data (which I can only seem to find dailies for what i'm interested in [forex])
Anyone know of any feeds I can play with? the quicker the better really, though no worries if prices are delayed by 15 minutes or whatever.
Thanks.

 

[obayesshelton]

I doubt you could get delayed streaming data for free. if you wanted delayed you would have to either download it on the fly and store it or find an API you can query.

Streaming market data is pretty expensive you would have to look at someone like morning star

All the options are going to cost you unless you go for something like yahoo etc

 

[Trader333]

What instruments are you wanting to have a datafeed for ?

 

[missluna1982]

What instruments are you wanting to have a datafeed for ?

hello

i have the same question, i need the opening price of the vix index in realtime, anyone got any ideas? looks like pandas can only access historical data from what i see.

 

[Lacanau]

What instruments are you wanting to have a datafeed for ?

Just the majors really. Spot Gold, Silver, and Crude would be nice too, but frankly I'm just looking for something to try out some algorithms on. Have you heard of anywhere that gives out any free feeds? Or even trials?

 

[drtro]

I doubt you could get delayed streaming data for free. if you wanted delayed you would have to either download it on the fly and store it or find an API you can query.

Streaming market data is pretty expensive you would have to look at someone like morning star

All the options are going to cost you unless you go for something like yahoo etc

There's plenty of free data sources out there. Google offers free intraday data.

http://www.quantshare.com/sa-426-6-...ntraday-and-tick-data-for-the-us-stock-market

 

[hhiusa]

There's plenty of free data sources out there. Google offers free intraday data.

http://www.quantshare.com/sa-426-6-...ntraday-and-tick-data-for-the-us-stock-market

You either pay for it from your brokerage firm or you buy the redistribution license $3,000/mo NYSE, which requires that you charge people access to your feed based upon non-professional and professional subscribers. Nothing is ever free.

**NYSE, NYSE Arca LLC, and NYSE MKT LLC reserve all rights to the securities information that Yahoo! Inc. makes available to you. You understand and acknowledge that such securities information does not reflect trading activity on markets other than NYSE, NYSE Arca, or NYSE MKT, as applicable, and are intended to provide you with a reference point only, rather than as a basis for making trading decisions. None of Yahoo! Inc., NYSE, NYSE Arca LLC, and NYSE MKT LLC guarantee such information nor shall any of them be liable for any loss due either to their negligence or to any cause beyond their reasonable control. Any redistribution of such information is strictly prohibited.

It is alright for historical data but most of the sites on the webpage don't allow you to go back that far. I won't be dropping my subscription anytime soon.

 

[bluk2015]

live data, use with python

So i've done a couple of courses on computational finance, and i'm pretty familiar with python so I want to start applying some of what i've learned on actual real time data feeds rather than downloading and computing historical data (which I can only seem to find dailies for what i'm interested in [forex])
Anyone know of any feeds I can play with? the quicker the better really, though no worries if prices are delayed by 15 minutes or whatever.
Thanks.

try to find the quantopian site, i am pretty sure it gets you what you are looking for. Not only is it free, if you develop a demonstrated profitable trading strategy funding is made available. Let me know whether this works for you. I am still stuck in learning python.
Bluk

 

[Lacanau]

try to find the quantopian site, i am pretty sure it gets you what you are looking for. Not only is it free, if you develop a demonstrated profitable trading strategy funding is made available. Let me know whether this works for you. I am still stuck in learning python.
Bluk

Thanks. I will look into it. What I have done was recreate a 'fake' data feed that feeds data to my program. I take historical data and pass it to the program which then recreates tick by tick market conditions. I won't go too far into it

If you're stuck learning python, I would recommend www.codecademy.com . Their python course is great and the functionality of the website makes learning the language a fast and above all enjoyable process.

 

[drtro]

You either pay for it from your brokerage firm or you buy the redistribution license $3,000/mo NYSE, which requires that you charge people access to your feed based upon non-professional and professional subscribers. Nothing is ever free.

**NYSE, NYSE Arca LLC, and NYSE MKT LLC reserve all rights to the securities information that Yahoo! Inc. makes available to you. You understand and acknowledge that such securities information does not reflect trading activity on markets other than NYSE, NYSE Arca, or NYSE MKT, as applicable, and are intended to provide you with a reference point only, rather than as a basis for making trading decisions. None of Yahoo! Inc., NYSE, NYSE Arca LLC, and NYSE MKT LLC guarantee such information nor shall any of them be liable for any loss due either to their negligence or to any cause beyond their reasonable control. Any redistribution of such information is strictly prohibited.

It is alright for historical data but most of the sites on the webpage don't allow you to go back that far. I won't be dropping my subscription anytime soon.

By now, I seriously can't tell if you are one well programmed troll bot or some stupid kid.

"Nothing is ever free."

Apparently ****posting is.

 

[hhiusa]

By now, I seriously can't tell if you are one well programmed troll bot or some stupid kid.

"Nothing is ever free."

Apparently ****posting is.

It takes two! You receive an e-mail whenever someone quotes you. That's not rocket science. I doesn't take long to post a reply from my phone.

 

[systematics]

I'm not sure if Quantopian offers Forex data? I though they only focused on stocks, ETFs, and indices?

If you are serious about trading and need a real-time data feed plus historical data - I would open an account with Interactive Brokers. As long as you meet their minimum account size requirement (which I think is $10k in US, should be similar in the UK - you will get access to whatever data you need, and Forex data is free from them.. incl. intra-day.

Also, if you are interested in Forex specifically - get a demo account from any broker that offers MetaTrade platform - they will offer historical forex data free of charge, you can download and store in text files on your computer. Just beware if you decide to trade through them, a lot of MetaTrader brokers are bucket shops... (not all though).

 

[techite]

You could begin testing with Yahoo Finance API.
I had recently uploaded a python program on github which is pretty simple to understand and may assist you in implementing your knowledge of computational finance over the variables provided by yahoo finance in .CSV format.
Have a look at https://github.com/techite/FinancialCalc/blob/master/StockDisplay.py

 

[Solas0077]

I'm not sure if Quantopian offers Forex data? I though they only focused on stocks, ETFs, and indices?

They have plans for that. There is another one, quantiacs that focuses exclusively on futures. Quantopian appears more popular. I am working on first algo since Price Action Lab recently added a feature for generating code for it.

Quantopian uses an algo to estimate impact of commission and partial fills. This is very good. I have tested some systems and I found a small effect in performance due to Quantopian estimated slippage but not very high.

There are some other services I have not tried like QuantConnect. I will be looking at quantiacs soon.