Show Posts
|
Pages: [1] 2 »
|
Hi Xiaoxiao,
I can make this script for you. I have some good experience with exchanges API's, bots and programming. Please pm me with details.
Depending on what functionality and how complex the script needs to be i can tell you the price.
|
|
|
You should definetely give more information about the strategy it is trading on. If it is an arbitrage strategy then selling the bot to multiple people will immeadiatly eliminate any potential profits, so the bot will become useless. Also it sounds like a big scam if you claim you make 5 BTC a month with almost no capital.
It would be better if you claimed the outperformance of the bot. So the bot makes 10%, the market did 5% for that period, hence the outperformance is 5%.
|
|
|
Search for the basis of your bot code on github. After this building the code for your own strategy is quit easy. Built a bot that does not trade to often, because paying the spread and transaction fees will eat your profit away.
|
|
|
I still do not agree with this partial fix of Bitstamp. If I place one order I want to be charged for just one order, regardless if it is bought into by three different orders. I decided to place one order so should be charged for just one. Bitstamp is just a scammer exchange with a bad reputation.
|
|
|
Thnx for your answers! Found it very interesting. I myself would be interested in the Excel sheet, as I have slightly different numbers than you do and am not absolutely sure how you did calculate the numbers.
For the strategy that I mentioned, the rate of return would be 56% on a yearly basis if you plug in the numbers from today (not sure if I did it correctly). Is this high in the bitcoin world? What was the highest you have seen? There must be some very big pitfalls? I have come by one pitfal, as of yesterday you can not register anymore on that exchange :s. Not sure what kind of exchange wouldn't allow people to register, but it shows that there could be many more flaws in the sites design.
|
|
|
Ok thnx.
So what would be wrong with buying 1 bitcoin in the market (564 at that time) and selling a futures contract to sell at 865 (at that time). You would lock in a tremendous rate. Also no "risk". What would stop you? and why do you need a fund size 4 times as large as the amount of bitcoins deployed?
|
|
|
Interesting read. You are obviously a very smart man.
I always thought that futures locked in the exchange rate, so how is it possible that this strategy loses in dollar terms?
|
|
|
The only real arbitrage I did was with bitcoinbuilder. I deposited money into Gox and got it out with around 70% profit margin. I was lucky to get it out in time, because two days later they froze their website. I don't see these opportunities anymore.
If you do arbitrage now, then if you calculate the arbitrage opportunities between exchanges after transaction costs, you can get on average 0.10$ per trade. So your just scalping the markets. This small profit margin wipes out very fast when you make some mistakes or do the wrong trade. Also you have to do alot and very fast to make it profitable.
|
|
|
When writing your bot in LISP or Scheme, can you connect to the websockets?
|
|
|
Hi there... I am running a platform where you can code your rule engine in JavaScript then test it and use it for real trading on BTC-e. There are some strategies already implemented and you can get a hang of how things work pretty fast. You can then use this knowledge to build your own bot You can find the platform here: http://www.cointelligence.netHave fun coding your own strategies and rules! Thnx!
|
|
|
One more question What are the rules for hitting an API. So basically if i hit an API for 1 minutes 200 times and then don't do it for 3.5 hours. Averaging once a second. Is that even allowed? Ím just trying to come up with ways to work around it.
|
|
|
Ok nice. Do traders sometimes end up with excess inventories/balances :p How big would that risk be?
|
|
|
Some exchanges have methods to check the status of an order. The most simple way is usually to check your funds. If you bought 1 btc and your btc funds are 1 btc higher than before => order filled.
Like you said, you cant hit the api continuously?, so how to work around this. For example: You usually check for the opportunity and one round trip (buy USD to BTC and immediately sell BTC to XXX) would look like this: Make one side of the trade Wait one sec Check balance if trade went through Wait one sec Make other side of the trade Wait one sec Check balance if trade went through Wait one sec
|
|
|
No need to reinvent the wheel. Take an existing bot and modify for your needs. I wasted several months on my own code until I dug deeper into what is out there. My code was usable,but still just the basics. With existing, you get the basics for free and you get to build something advanced more quickly.
For nodejs I run Gekko. Search for it here or Github.
Yess thats a good idea. Thanks for the pointer, I will take a first look at Gekko. So I was wondering.. What happens when you see a trading opportunity? Do you place a limit order at one of the open bid-ask prices or do you come in with a market order? How do you check if the trade was really executed?
|
|
|
Hi,
I got kind of inspired by all the trading bots offered on this site and now I want to build my own bot. Most of the bots that I have seen here are more user friendly, with a nice interface ect. I think the best way to build my own bot is just some plain script that does the trading and nothing more. I just want to do it myself to get to know the inside outs of trading bots and that I am not reliable to someone else for further improvements. I need some pointers in the right direction and am willing to learn some new skills to do this. At the moment it needs to update and search for trading opportunities every 1 to 5 seconds, but maybe in 1 year time this needs to be considerably faster. So im looking to build something that is scalable on speed, as it becomes harder and harder to beat other players.
At the moment I was thinking to write the code in Node js and javascript. How does this sound? I am collecting trading data in a mysql database and can do the computations for trading opportunities there. The reason for this is, that I think that this method is less prone to errors. For example, exchange sites api lag sometimes and I need to collect data for when a trade is confirmed ect. Is it better?
Are there any other languages that I could best write my code in?
Thanks and help will be appreciated.
|
|
|
Hi there, I need some help accessing the huobi api. http://market.huobi.com/staticmarket/detail.htmlin PHP $obj = { "sells": [ { "price": 3840, "level": 0, "amount": 1 }, { "price": "3840.05", "level": 0, "amount": 0.287 }, { "price": 3841, "level": 0, "amount": 0.1 } ], "p_new": 3838, "level": -72.12, "amount": 82792, "total": 321774060.34653, "amp": -2, "p_open": 3910.12, "p_high": 3925, "p_low": 3809.99, "p_last": 3910.12 } echo "Ask " . $obj->sells[0]->price; // does not work echo "Volume" . $obj->amount;// does not work Help will be appreciated.
|
|
|
Hi there,
I need some help accessing the huobi api. in PHP
$obj = { "sells": [ { "price": 3840, "level": 0, "amount": 1 }, { "price": "3840.05", "level": 0, "amount": 0.287 }, { "price": 3841, "level": 0, "amount": 0.1 } ], "p_new": 3838, "level": -72.12, "amount": 82792, "total": 321774060.34653, "amp": -2, "p_open": 3910.12, "p_high": 3925, "p_low": 3809.99, "p_last": 3910.12 } echo "test1 " . $obj->sells[0]->price; echo "test2 " . $obj->amount;
|
|
|
The buy/sell wall is now checking for every 3000 open orders? Can you make the buy/sell wall calculated differently? So for example check all open bid orders above $400, or not more than 30% away from the midpoint (bid/ask/2), same for open asks order.
Just to avoid counting open orders of buying 10.000 bitcoin for 0.01$ in the orderbook ect.
|
|
|
|