增加交易策略、交易指标、量化库代码等文件夹

1.交易策略/1.CTA策略/1.松鼠策略/6.松鼠SF15_随机小波段策略/使用文稿/VNPY/vip15.py

@@ -0,0 +1,155 @@
+from vnpy_ctastrategy import (
+    CtaTemplate,
+    TargetPosTemplate,
+    StopOrder,
+    TickData,
+    BarData,
+    TradeData,
+    OrderData,
+    BarGenerator,
+    ArrayManager,
+)
+import numpy as np
+import numpy as np
+import talib
+
+
+class vip15(CtaTemplate):
+    author = "quant789.com"
+
+    period = 5
+    pds = 24
+    trailing_stop_rate = 65
+    lots = 1
+    kg = 0
+    parameters = ["period", "pds", "trailing_stop_rate", "lots"]
+    variables = ["sdo_value", "lots", "barcout", "kg", "dd", "ss"]
+
+    def __init__(self, cta_engine, strategy_name, vt_symbol, setting):
+        super().__init__(cta_engine, strategy_name, vt_symbol, setting)
+
+        self.bg = BarGenerator(self.on_bar)
+        self.am = ArrayManager(300)  # Adjust size as needed
+
+        self.sdo_array = 0
+        self.barcout = 0
+
+        self.dd = 0
+        self.ss = 0
+
+        self.over_bought = 40
+        self.over_sold = -40
+        self.lb_period = 50
+
+        self.highest_low_after_entry = 0
+        self.lowest_high_after_entry = 0
+        self.liq_ka = 1
+        self.dliq_point = 0
+        self.kliq_point = 0
+        self.ent_bar = 0
+        self.ent_price = 0
+
+    def calculate_sdo(self):
+        if len(self.am.close_array) < self.lb_period:
+            return []
+
+        close = self.am.close_array
+        dist = np.abs(close - np.roll(close, self.period))
+        lowest_dist = np.minimum.accumulate(dist[::-1])[::-1]
+        highest_dist = np.maximum.accumulate(dist[::-1])[::-1]
+        
+        # 避免除以零
+        denom = highest_dist - lowest_dist
+        denom[denom == 0] = 1e-10  # 将零值替换为一个很小的数
+        dval = np.where(denom != 0, (dist - lowest_dist) / denom, 0)
+        
+        ddval = np.zeros_like(dval)
+        ddval[close > np.roll(close, self.period)] = dval[close > np.roll(close, self.period)]
+        ddval[close < np.roll(close, self.period)] = -dval[close < np.roll(close, self.period)]
+        
+        sdo = np.convolve(ddval, np.ones(self.pds), 'valid') / self.pds * 100
+        return sdo
+
+    def on_init(self):
+        self.write_log("策略初始化")
+        self.load_bar(10)
+
+    def on_start(self):
+        self.write_log("策略启动")
+
+    def on_stop(self):
+        self.write_log("策略停止")
+
+    def on_tick(self, tick: TickData):
+        self.bg.update_tick(tick)
+
+    def on_bar(self, bar: BarData):
+        self.am.update_bar(bar)
+        if not self.am.inited:
+            return
+
+        self.sdo_array = self.calculate_sdo()
+        
+        cond1 = self.sdo_array[-2] > self.over_bought and self.sdo_array[-1] <= self.over_bought
+        cond2 = self.sdo_array[-2] < self.over_sold and self.sdo_array[-1] >= self.over_sold
+
+        hh = self.am.high_array[-self.period:].max()
+        ll = self.am.low_array[-self.period:].min()
+        
+        if cond1==True or cond2==True:
+            self.dd = hh
+            self.ss = ll
+            self.kg = 1
+            self.barcout = 0
+
+        if self.kg > 0:
+            self.barcout += 1
+
+        if self.barcout > self.pds:
+            self.barcout = 0
+            self.dd = 0
+            self.ss = 0
+            self.kg = 0
+        
+        if  self.kg > 0 and bar.high_price >= self.dd and self.dd>0 : #
+            self.buy(bar.close_price, self.lots)
+            self.kg = 0
+            self.barcout = 0
+            self.highest_low_after_entry = bar.close_price
+            self.initialize_vars_on_entry(bar)
+        if  self.kg > 0 and bar.low_price <= self.ss and self.ss>0 : #
+            self.short(bar.close_price, self.lots)
+            self.kg = 0
+            self.barcout = 0
+            self.lowest_high_after_entry = bar.close_price
+            self.initialize_vars_on_entry(bar)
+        else:
+            self.trailing_stop(bar)
+
+        self.put_event()
+
+        
+
+    def trailing_stop(self, bar: BarData):
+        if self.pos > 0:
+            self.highest_low_after_entry = max(self.highest_low_after_entry, bar.low_price)
+            self.dliq_point = self.highest_low_after_entry - (bar.open_price * self.trailing_stop_rate / 10000) * self.liq_ka
+            if bar.low_price <= self.dliq_point and bar.datetime.timestamp() > self.ent_bar:
+                self.sell(bar.close_price , abs(self.pos))
+        elif self.pos < 0:
+            self.lowest_high_after_entry = min(self.lowest_high_after_entry, bar.high_price)
+            self.kliq_point = self.lowest_high_after_entry + (bar.open_price * self.trailing_stop_rate / 10000) * self.liq_ka
+            if bar.high_price >= self.kliq_point and bar.datetime.timestamp() > self.ent_bar:
+                self.cover(bar.close_price , abs(self.pos))
+
+    def initialize_vars_on_entry(self, bar: BarData):
+        self.ent_bar = bar.datetime.timestamp()
+        self.ent_price = bar.close_price
+        self.liq_ka = 1
+
+
+    def on_order(self, order: OrderData):
+        pass
+
+    def on_stop_order(self, stop_order: StopOrder):
+        pass