Abnormal Volume Effect in the Stock Market

1import numpy as np
2from AlgorithmImports import *
3from typing import List, Dict
4from pandas.core.frame import DataFrame
5class AbnormalVolumeEffectStockMarket(QCAlgorithm):
6def Initialize(self) -> None:
7self.SetStartDate(2000, 1, 1)
8self.SetCash(100000)
9market:Symbol = self.AddEquity('SPY', Resolution.Daily).Symbol
10
11self.data:Dict[Symbol, SymbolData] = {}
12self.period:int = 66
13self.leverage:int = 5
14self.std_treshold:float = 2.33
15self.performance_treshold:float = 0.01
16
17self.long:List[Symbol] = []
18
19self.selection_flag:bool = False
20self.last_selection:List[Fundamental] = []
21
22self.fundamental_sorting_key = lambda x: x.MarketCap
23self.fundamental_count:int = 1000
24self.UniverseSettings.Resolution = Resolution.Daily
25self.AddUniverse(self.FundamentalSelectionFunction)
26self.Settings.MinimumOrderMarginPortfolioPercentage = 0.
27self.settings.daily_precise_end_time = False
28self.Schedule.On(self.DateRules.MonthStart(market), self.TimeRules.AfterMarketOpen(market), self.Selection)
29
30def OnSecuritiesChanged(self, changes: SecurityChanges) -> None:
31for security in changes.AddedSecurities:
32    security.SetFeeModel(CustomFeeModel())
33    security.SetLeverage(self.leverage)
34
35def FundamentalSelectionFunction(self, fundamental: List[Fundamental]) -> List[Symbol]:
36# Update the rolling window every day.
37for stock in fundamental:
38    symbol:Symbol = stock.Symbol
39    if symbol in self.data:
40        # Store daily price and volume.
41        self.data[symbol].update(stock.AdjustedPrice, stock.Volume)
42
43# return already selected universe during the month
44if self.selection_flag:
45    self.selection_flag = False
46    selected:List[Fundamental] = [
47        x for x in fundamental if x.HasFundamentalData and x.Market == 'usa' and x.MarketCap != 0
48    ]
49    if len(selected) > self.fundamental_count:
50        selected = sorted(selected, key = self.fundamental_sorting_key, reverse=True)[:self.fundamental_count]
51    self.last_selection = selected
52    # Warmup price rolling windows.
53    for stock in self.last_selection:
54        symbol:Symbol = stock.Symbol
55        if symbol in self.data:
56            continue
57        self.data[symbol] = SymbolData(symbol, self.period, -1)
58        history:DataFrame = self.History(symbol, self.period, Resolution.Daily)
59        if history.empty:
60            self.Log(f"Not enough data for {symbol} yet")
61            continue
62        if 'close' in history and 'volume' in history:
63            closes:Series = history.loc[symbol]['close']
64            volumes:Series = history.loc[symbol]['volume']
65            for (time1, close), (time2, volume) in zip(closes.items(), volumes.items()):
66                self.data[symbol].update(close, volume)
67# fundamental returned ready data
68for stock in self.last_selection:
69    symbol:Symbol = stock.Symbol
70    if symbol not in self.data:
71        continue
72    if not self.data[symbol].is_ready():
73        continue
74    volumes:List[float] = [x for x in self.data[symbol]._volume]
75    volume_mean:float = np.mean(volumes)
76    volume_std:float = np.std(volumes)
77    volume:float = volumes[0] # Takes todays volume
78       
79    closes:List[float] = [x for x in self.data[symbol]._price][:2] # First two are newest
80    todays_return:float = (closes[0] - closes[1]) / closes[1]
81     
82    if volume > volume_mean + (self.std_treshold * volume_std):
83        # selects only firms with no abnormal volume over the preceding 30 trading days
84        if (self.data[symbol]._abnormal_date == -1) or (self.data[symbol]._abnormal_date < (self.Time - timedelta(days=30))):
85            # if the stocks finished the day with at least a 1% gain
86            if todays_return >= self.performance_treshold:
87                self.long.append(symbol)
88        self.data[symbol]._abnormal_date = self.Time
89            
90return self.long
91
92def OnData(self, data: Slice) -> None:
93# Trade execution
94targets:List[PortfolioTarget] = [PortfolioTarget(symbol, 1. / len(self.long)) for symbol in self.long if symbol in data and data[symbol]]
95self.SetHoldings(targets, True)
96self.long.clear()
97
98def Selection(self) -> None:
99self.selection_flag = True
100
101class CustomFeeModel(FeeModel):
102def GetOrderFee(self, parameters):
103fee = parameters.Security.Price * parameters.Order.AbsoluteQuantity * 0.00005
104return OrderFee(CashAmount(fee, "USD"))
105
106class SymbolData():
107def __init__(self, symbol:Symbol, period:int, abnormal_date:datetime):
108self._symbol:Symbol = symbol
109self._price:RollingWindow = RollingWindow[float](period)
110self._volume:RollingWindow = RollingWindow[float](period)
111self._abnormal_date:datetime = abnormal_date
112
113def update(self, price:float, volume:float):
114self._price.Add(price)
115self._volume.Add(volume)
116
117def is_ready(self) -> bool:
118return self._price.IsReady and self._volume.IsReady