期限套利策略，要能够稳定获得各币种期货和现货的报价，计算各币种的基差数据，然后当基差突破阈值时触发交易。以下代码可以得到稳定报价，计算基差数据，然后当突破阈值时收到提示邮件（还没写好交易模块）。

这里记录一下，开发过程遇到的主要问题，以及如何解决的，如果不感兴趣，可以跳过，直接看代码。

问题一：websocket报价数据不更新

简单的ws.run_forever，通常在运行两三天后，会出现报价卡在某个时点，不能更新的情况。例如，现货报价停留在2021年12月25日11:00。

解决方案：用while true + try except语句，一旦websock链接有问题，就发送邮件通知（不知道为啥，一直没有收到报错邮件，发送邮件的代码可能有问题），回收内存，再重新链接。

    def s_start(self):while True:try:s_ws=WebSocketApp(url='wss://stream.binance.com:9443/ws',on_open=self._s_on_open,on_message=self._s_on_message,on_error=self._s_on_error,on_close=self._s_on_close)s_ws.run_forever() except Exception as e:content="%s the spot webscoket is broken,check if it restart"%eself._email(content)gc.collect()

问题二：启动阶段报价数据为空

因为涉及到多币种，刚刚建立链接的时候，报价数据获取有先后，有可能btc数据有了，但是bnb数据要到10秒甚至20秒后才能收到。这样程序会因报错而中断。

解决方案：定义好报错的类型，用@retry装饰器，保证程序在遇到特定类型报错的时候，能够持续运行下去。

    def retry_if_index_error(exception):return isinstance(exception,IndexError)@retry(retry_on_exception=retry_if_index_error)def get_basis(self):

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Sun Dec  5 20:15:41 2021
It seems to work really good, the price and basis data is correct after running 13 days.
2021/12/13 16:30 begin
@author: yszhu
"""from websocket import WebSocketApp
import json,threading
import pysnooper
import time
import smtplib
from email.mime.text import MIMEText
from retrying import retry
import logging
import gcclass basis_dif(object):"""create variable save dilivery contract price,spot price and basis difference"""def __init__(self,contract_date=220325):#delivery contract priceself.d_eth_price=Noneself.d_eth_time=Noneself.d_btc_price=Noneself.d_btc_time=Noneself.d_bnb_price=Noneself.d_bnb_price=None#spot priceself.eth_price=Noneself.eth_time=Noneself.btc_price=Noneself.btc_time=Noneself.bnb_price=Noneself.bnb_time=None#delivery and spot basis differenceself.btc_basis_dif=Noneself.eth_basis_dif=Noneself.bnb_basis_dif=Noneself.contract_date=contract_dateself.basis_dif_threshold=0.035#emailself._mail_host="smtp.qq.com"self._mail_user="315960451"self._mail_pwd="brvjltflaofrbhcb"self._sender="youradress@qq.com"self._receivers="youradress@qq.com"# I use self._break_threshold to avoid sending email repeatedly.# when self._break_threshold = True ,it means the basis diffrence# now is greater than the threshold,so if the diffence becomes smaller# than the threshold, this is the first time of break during one operation period.# So ,I will receive the email ,and then operate my account.self._break_threshold = True#websocket for delivery contrat price def _d_on_open(self,d_ws):data={"method":"SUBSCRIBE","params":["btcusd_%s@aggTrade"%self.contract_date,"ethusd_%s@aggTrade"%self.contract_date,"bnbusd_%s@aggTrade"%self.contract_date],"id": 1}                        d_ws.send(json.dumps(data))def _d_on_message(self,d_ws,d_msg):d_msg=json.loads(d_msg)if 's' in d_msg and d_msg['s']=="BTCUSD_%s"%self.contract_date:self.d_btc_price=float(d_msg['p'])self.d_btc_time=d_msg['T']self.d_btc_time=int(self.d_btc_time/1000)self.d_btc_time=time.localtime(self.d_btc_time)self.d_btc_time=time.strftime("%Y-%m-%d %H:%M:%S",self.d_btc_time)if 's' in d_msg and d_msg['s']=="ETHUSD_%s"%self.contract_date:self.d_eth_price=float(d_msg['p'])self.d_eth_time=d_msg['T']self.d_eth_time=int(self.d_eth_time/1000)self.d_eth_time=time.localtime(self.d_eth_time)self.d_eth_time=time.strftime("%Y-%m-%d %H:%M:%S",self.d_eth_time)if 's' in d_msg and d_msg['s']=="BNBUSD_%s"%self.contract_date:self.d_bnb_price=float(d_msg['p'])self.d_bnb_time=d_msg['T']self.d_bnb_time=int(self.d_bnb_time/1000)self.d_bnb_time=time.localtime(self.d_bnb_time)self.d_bnb_time=time.strftime("%Y-%m-%d %H:%M:%S",self.d_bnb_time)def _d_on_close(self,d_ws):print("##connection closed##")def _d_on_error(self,d_ws,error):print(f"on error:{error}")def d_start(self):while True:try:d_ws=WebSocketApp(url='wss://dstream.binance.com/ws',on_open=self._d_on_open,on_message=self._d_on_message,on_error=self._d_on_error,on_close=self._d_on_close)d_ws.run_forever()except Exception as e:content="%s the future webscoket is broken,check if it restart"%eself._email(content)gc.collect()#websocket for spot price def _s_on_open(self,s_ws):data={"method": "SUBSCRIBE","params": ["btcusdt@aggTrade","ethusdt@aggTrade","bnbusdt@aggTrade"],"id": 2}               s_ws.send(json.dumps(data))def _s_on_message(self,s_ws,s_msg):s_msg=json.loads(s_msg)if 's' in s_msg and s_msg['s']=="BTCUSDT":self.btc_price=float(s_msg['p'])self.btc_time=s_msg['T']self.btc_time=int(self.btc_time/1000)self.btc_time=time.localtime(self.btc_time)self.btc_time=time.strftime("%Y-%m-%d %H:%M:%S",self.btc_time)if 's' in s_msg and s_msg['s']=="ETHUSDT":self.eth_price=float(s_msg['p'])self.eth_time=s_msg['T']self.eth_time=int(self.eth_time/1000)self.eth_time=time.localtime(self.eth_time)self.eth_time=time.strftime("%Y-%m-%d %H:%M:%S",self.eth_time)if 's' in s_msg and s_msg['s']=="BNBUSDT":self.bnb_price=float(s_msg['p'])self.bnb_time=s_msg['T']self.bnb_time=int(self.bnb_time/1000)self.bnb_time=time.localtime(self.bnb_time)self.bnb_time=time.strftime("%Y-%m-%d %H:%M:%S",self.bnb_time)def _s_on_close(self,s_ws):print("##connection closed##")def _s_on_error(self,s_ws,error):print(f"on error:{error}")def s_start(self):while True:try:s_ws=WebSocketApp(url='wss://stream.binance.com:9443/ws',on_open=self._s_on_open,on_message=self._s_on_message,on_error=self._s_on_error,on_close=self._s_on_close)s_ws.run_forever() except Exception as e:content="%s the spot webscoket is broken,check if it restart"%eself._email(content)gc.collect()   #because there are 7 kind of coin with spot and future price , so at#the begining , there maybe no data for self.bnb_price for the lack of liquidity,#In this case , python will raise IndexError#we need trying when TypeError is raised. def retry_if_index_error(exception):return isinstance(exception,IndexError)@retry(retry_on_exception=retry_if_index_error)def get_basis(self):while True:self.btc_basis_dif = self.d_btc_price/self.btc_price-1self.eth_basis_dif = self.d_eth_price/self.eth_price-1self.bnb_basis_dif = self.d_bnb_price/self.bnb_price-1print("btc_basis_dif is %f" % self.btc_basis_dif)print("btc_d_price is %f %s"%(self.d_btc_price,self.d_btc_time))print("btc_s_price is %f %s"%(self.btc_price,self.btc_time))print("eth_basis_dif is %f"%self.eth_basis_dif)print("eth_d_price is %f %s"%(self.d_eth_price,self.d_eth_time))print("eth_s_price is %f %s"%(self.eth_price,self.eth_time))print("bnb_basis_dif is %f"%self.bnb_basis_dif)print("bnb_d_price is %f %s"%(self.d_bnb_price,self.d_bnb_time))print("bnb_s_price is %f %s"%(self.bnb_price,self.bnb_time))basis_dif_dict={"btc":[self.btc_basis_dif,self.btc_price,self.d_btc_price],"eth":[self.eth_basis_dif,self.eth_price,self.d_eth_price],"bnb":[self.bnb_basis_dif,self.bnb_price,self.d_bnb_price],}basis_dif_dict=sorted(basis_dif_dict.items(),key=lambda x:x[1],reverse=True)greatest_basis_dif=basis_dif_dict[0][1][0]        print("the greatest basis is %s %f,Spot pirice %f,future price %f"%(basis_dif_dict[0][0],greatest_basis_dif,basis_dif_dict[0][1][1],basis_dif_dict[0][1][2]))if greatest_basis_dif>self.basis_dif_threshold:if self._break_threshold == True:content="the greatest basis is %s %f,Spot pirice %f,future price %f"%(basis_dif_dict[0][0],greatest_basis_dif,basis_dif_dict[0][1][1],basis_dif_dict[0][1][2])self._email(content)self._break_threshold = Falseif greatest_basis_dif<self.basis_dif_threshold:self._break_threshold= Truedef _email(self,content):'''if the basis_dif reached the threshold, send an emailparam:content: which coin's bais_dif has reach the threshold at what price and when'''message=MIMEText(content,'plain',"utf-8")message['Subject']="from new greatest basis dif:basis difference reached the threshold"message['From']=self._sendermessage['to']=self._receivers[0]smtpObj=smtplib.SMTP()smtpObj.connect(self._mail_host,25)smtpObj.login(self._mail_user,self._mail_pwd)smtpObj.sendmail(self._sender,self._receivers,message.as_string())smtpObj.quit()if __name__=="__main__":test=basis_dif()# because I want to recieve spot and contract price at the same time, so I create two threads t1=threading.Thread(target=test.d_start)t2=threading.Thread(target=test.s_start)t1.start()t2.start()test.get_basis()

binance 币本位交割合约期限套利交易（三）--获取基差数据相关推荐

【币圈小试牛刀】期现套利（上）——期限套利原理
大家好!我是币圈玖哥,匆匆忙忙进场,匆匆忙忙进厂-提醒本文不构成任何投资建议,只是作为学习分享,本期给大家带来的策略是:期限套利,套取永续合约的资金费率,由于风险偏低,适合新手使用! 文章目录 1.期 ...
如何规避期货套利交易的陷阱
我们都知道在套利交易中,就风险来说,相比单边交易风险要小得多:但是,尽管如此也不能放松对套利交易风险的控制,如果操作不当,套利交易也会造成滚雪球效应,损失也会一步一步扩大.那么如何规避套利交易的一些陷 ...
什么是套利？什么是套利交易
什么是套利?套利是什么意思? 套利(Interest Arbitrage) 套利概述在一般情况下,西方各个国家的利息率的高低是不相同的,有的国家利息率较高,有的国家利息率较低.利息率高低是国际资 ...
做空不再瞻前顾后！CoinBene满币USDT永续合约实盘上线(1)
CoinBene满币 USDT永续合约体验争霸赛吸引了众多合约玩家参与其中,共计27位用户赢得2340USDT现金大奖.颁奖完成后,CoinBene满币正式上线USDT永续合约实盘交易,USDT永续合 ...
交割合约(期货合约)是什么?
交割合约又称期货合约,是指在未来特定时间(交割日)以预定价格买卖标的资产的协定,是一种金融衍生品.与永续合约不同,交割合约会到期,有确定的到期日/交割日.当交割合约到期时,无论标的资产的实际价格如何, ...
开仓风险计算器.xlsx（可计算：名义价值、最大资金亏损、开仓所需保证金、开仓资金杠杆、最小逐仓保证金、U本位需开张数、币本位需开张数）
这个excel版的开仓风险计算器我一年前就写好了,一直在用,后面还迭代了些新功能,如今的计算应该能满足绝大部分普通交易者的需求了. 不废话,上图: ↓ 点击卡片下载计算器 ↓ 开仓风险计算器.xlsx ...
股指期货的理论价格与期限套利
股指期货的定价利用的是无风险套利原理,也就是说股指期货的价格应当消除无风险套利的机会,否则就会有人进行套利.对于一般的投资者来说,只要了解股指期货价格与现货指数.无风险利率.红利率.到期前时间长短有关 ...
如何运用债市杠杆套利交易博取收益
如何运用债市杠杆套利交易博取收益套利交易策略在目前我国的固定收益市场中,合理运用套利交易策略可以获得超额收益,主要包括以下套利交易策略. 1)息差交易策略:指通过不断正回购融资并买入债券的策略,只 ...
大所交割合约到期情况观察
ok交割合约到期时间本季度:20210924(9月最后一个周五) 但策略运行到9月20日12点左右就停止了 9月22日重新运行后,out文件时间刷新,但内容无法刷新发现是私钥没上传上传后,重新运 ...

binance 币本位交割合约期限套利交易（三）--获取基差数据

问题一：websocket报价数据不更新

问题二：启动阶段报价数据为空

binance 币本位交割合约期限套利交易（三）--获取基差数据相关推荐

最新文章

热门文章