add real time trading and intraday examples

jaqs/example/eventdriven/DoubleMA.py

@@ -0,0 +1,165 @@
+# encoding: utf-8
+
+import time
+
+import numpy as np
+
+from jaqs.trade import common
+from jaqs.trade.strategy import EventDrivenStrategy
+from jaqs.data.dataservice import RemoteDataService
+from jaqs.trade import model
+from jaqs.trade.realtime import EventRealTimeInstance
+from jaqs.trade.backtest import EventBacktestInstance
+from jaqs.trade.tradegateway import RealTimeTradeApi, BacktestTradeApi
+from jaqs.trade.portfoliomanager import PortfolioManager
+import jaqs.util as jutil
+import jaqs.trade.analyze.analyze as ana
+
+result_dir_path = jutil.join_relative_path('../output/double_ma')
+is_backtest = True
+
+
+class DoubleMaStrategy(EventDrivenStrategy):
+    """"""
+    def __init__(self):
+        super(DoubleMaStrategy, self).__init__()
+        self.symbol = ''
+
+        self.fast_ma_len = 13
+        self.slow_ma_len = 23
+
+        self.window_count = 0
+        self.window = self.slow_ma_len + 1
+
+        self.price_arr = np.zeros(self.window)
+        self.fast_ma = 0
+        self.slow_ma = 0
+        self.pos = 0
+
+        self.buy_size_unit = 1
+        self.output = True
+
+    def init_from_config(self, props):
+        super(DoubleMaStrategy, self).init_from_config(props)
+        self.symbol = props.get('symbol')
+        self.init_balance = props.get('init_balance')
+
+    def buy(self, quote, size=1):
+        if hasattr(quote, 'bidprice1'):
+            ref_price = (quote.bidprice1 + quote.askprice1) / 2.0
+        else:
+            ref_price = quote.close
+
+        task_id, msg = self.ctx.trade_api.place_order(quote.symbol, common.ORDER_ACTION.BUY, ref_price + 3, self.buy_size_unit * size)
+        if (task_id is None) or (task_id == 0):
+            print("place_order FAILED! msg = {}".format(msg))
+
+    def sell(self, quote, size=1):
+        if hasattr(quote, 'bidprice1'):
+            ref_price = (quote.bidprice1 + quote.askprice1) / 2.0
+        else:
+            ref_price = quote.close
+
+        task_id, msg = self.ctx.trade_api.place_order(quote.symbol, common.ORDER_ACTION.SHORT, ref_price - 3, self.buy_size_unit * size)
+        if (task_id is None) or (task_id == 0):
+            print("place_order FAILED! msg = {}".format(msg))
+
+    def on_tick(self, quote):
+        if hasattr(quote, 'bidprice1'):
+            mid = (quote.bidprice1 + quote.askprice1) / 2.0
+        else:
+            mid = quote.close
+        self.price_arr[0: self.window - 1] = self.price_arr[1: self.window]
+        self.price_arr[-1] = mid
+        self.window_count += 1
+
+        if self.window_count <= self.window:
+            return
+
+        self.fast_ma = np.mean(self.price_arr[-self.fast_ma_len - 1:])
+        self.slow_ma = np.mean(self.price_arr[-self.slow_ma_len - 1:])
+
+        print(quote)
+        print("Fast MA = {:.2f}     Slow MA = {:.2f}".format(self.fast_ma, self.slow_ma))
+        if self.fast_ma > self.slow_ma:
+            if self.pos == 0:
+                self.buy(quote, 1)
+
+            elif self.pos < 0:
+                self.buy(quote, 2)
+
+        elif self.fast_ma < self.slow_ma:
+            if self.pos == 0:
+                self.sell(quote, 1)
+            elif self.pos > 0:
+                self.sell(quote, 2)
+
+    def on_quote(self, quote_dic):
+        quote = quote_dic.get(self.symbol)
+        self.on_tick(quote)
+
+    def on_trade(self, ind):
+        print "\nStrategy on trade: "
+        print(ind)
+        self.pos = self.ctx.pm.get_pos(self.symbol)
+
+    def on_order_status(self, ind):
+        if self.output:
+            print "\nStrategy on order status: "
+            print(ind)
+
+    def on_task_status(self, ind):
+        if self.output:
+            print "\nStrategy on task ind: "
+            print(ind)
+
+
+def run_strategy():
+    if is_backtest:
+        props = {"symbol": "rb1710.SHF",
+                 "start_date": 20170510,
+                 "end_date": 20170930,
+                 "bar_type": "1M",  # '1d'
+                 "init_balance": 2e4}
+
+        tapi = BacktestTradeApi()
+        ins = EventBacktestInstance()
+
+    else:
+        props = {'symbol': 'rb1801.SHF'}
+        tapi = RealTimeTradeApi()
+        ins = EventRealTimeInstance()
+
+    tapi.use_strategy(8)
+
+    ds = RemoteDataService()
+    strat = DoubleMaStrategy()
+    pm = PortfolioManager()
+
+    context = model.Context(data_api=ds, trade_api=tapi, instance=ins,
+                            strategy=strat, pm=pm)
+
+    ins.init_from_config(props)
+    if not is_backtest:
+        ds.subscribe(props['symbol'])
+
+    ins.run()
+    if not is_backtest:
+        time.sleep(9999)
+    ins.save_results(folder_path=result_dir_path)
+
+
+def analyze():
+    ta = ana.EventAnalyzer()
+
+    ds = RemoteDataService()
+    ds.init_from_config()
+
+    ta.initialize(data_server_=ds, file_folder=result_dir_path)
+
+    ta.do_analyze(result_dir=result_dir_path, selected_sec=[])
+
+
+if __name__ == "__main__":
+    run_strategy()
+    analyze()

jaqs/example/eventdriven/IntradaySpread.py

@@ -0,0 +1,123 @@
+# encoding: utf-8
+
+import numpy as np
+
+from jaqs.trade.strategy import EventDrivenStrategy
+from jaqs.trade import common, model
+
+from jaqs.data.dataservice import RemoteDataService
+from jaqs.trade.backtest import EventBacktestInstance
+from jaqs.trade.tradegateway import BacktestTradeApi
+from jaqs.trade.portfoliomanager import PortfolioManager
+import jaqs.util as jutil
+import jaqs.trade.analyze.analyze as ana
+
+result_dir_path = jutil.join_relative_path('../output/calendar_spread')
+
+
+class SpreadStrategy(EventDrivenStrategy):
+    """"""
+    def __init__(self):
+        super(SpreadStrategy, self).__init__()
+        self.symbol = ''
+
+        self.s1 = ''
+        self.s2 = ''
+        self.quote1 = None
+        self.quote2 = None
+
+        self.window = 8
+        self.idx = -1
+        self.spread_arr = np.empty(self.window, dtype=float)
+
+        self.threshold = 6.5
+
+    def init_from_config(self, props):
+        super(SpreadStrategy, self).init_from_config(props)
+        self.symbol = props.get('symbol')
+        self.init_balance = props.get('init_balance')
+
+        self.s1, self.s2 = self.symbol.split(',')
+
+    def on_cycle(self):
+        pass
+
+    def long_spread(self, quote1, quote2):
+        self.ctx.trade_api.place_order(quote1.symbol, common.ORDER_ACTION.BUY, quote1.close + 1, 1)
+        self.ctx.trade_api.place_order(quote2.symbol, common.ORDER_ACTION.SELL, quote2.close - 1, 1)
+
+    def short_spread(self, quote1, quote2):
+        self.ctx.trade_api.place_order(quote2.symbol, common.ORDER_ACTION.BUY, quote2.close + 1, 1)
+        self.ctx.trade_api.place_order(quote1.symbol, common.ORDER_ACTION.SELL, quote1.close - 1, 1)
+
+    def on_quote(self, quote_dic):
+        q1 = quote_dic.get(self.s1)
+        q2 = quote_dic.get(self.s2)
+        self.quote1 = q1
+        self.quote2 = q2
+
+        if self.quote1.time > 142800 and self.quote1.time < 200000:
+            # self.cancel_all_orders()
+            # self.liquidate(self.quote1, 3, tick_size=1.0, pos=self.ctx.pm.get_pos(self.quote1.symbol))
+            # self.liquidate(self.quote2, 3, tick_size=1.0, pos=self.ctx.pm.get_pos(self.quote2.symbol))
+            # return
+            pass
+
+        spread = q1.close - q2.close
+        self.idx += 1
+        self.spread_arr[self.idx % self.window] = spread
+
+        if self.idx <= self.window:
+            return
+
+        mean = self.spread_arr.mean()
+        if (spread - mean) > self.threshold:
+            self.short_spread(q1, q2)
+        elif (spread - mean) < -self.threshold:
+            self.long_spread(q1, q2)
+
+    def on_trade(self, ind):
+        print(ind)
+        print(self.ctx.pm.get_position(ind.symbol))
+
+
+props = {
+    "symbol"                : "rb1801.SHF,hc1801.SHF",
+    "start_date"            : 20170801,
+    "end_date"              : 20170910,
+    "bar_type"              : "1M",
+    "init_balance"          : 3e4,
+    "commission_rate": 2E-4
+}
+
+
+def run_strategy():
+    tapi = BacktestTradeApi()
+    ins = EventBacktestInstance()
+
+    ds = RemoteDataService()
+    strat = SpreadStrategy()
+    pm = PortfolioManager()
+
+    context = model.Context(data_api=ds, trade_api=tapi, instance=ins,
+                            strategy=strat, pm=pm)
+
+    ins.init_from_config(props)
+    ins.run()
+    ins.save_results(folder_path=result_dir_path)
+
+
+def analyze():
+    ta = ana.EventAnalyzer()
+
+    ds = RemoteDataService()
+    ds.init_from_config()
+
+    ta.initialize(data_server_=ds, file_folder=result_dir_path)
+
+    ta.do_analyze(result_dir=result_dir_path, selected_sec=props['symbol'].split(','))
+
+
+if __name__ == "__main__":
+    run_strategy()
+    analyze()