Timeouts and optimization

I want to implement a very straightforward strategy:

”Once a day, locate 10 stocks (out of all US stocks) with highest price change during last 100 days”.

To my understanding, it means I need to maintain a rolling window of trade bars gor each US stock, and inside “OnData” - when I get a daily bar - to add it to relevant rolling window and then to calculate a change for each stock and then submit needed orders.

The “coarse filter” function needs to return all stocks from US market + all stocks I currently invested in (otherwise they will be removed from the universe and I will stop get trading bars for them in “OnData”).

Does everything described above makes sense or there is a better alternative? In my backtests everything works (arguably not fast as I would like to but I can live with that) but during live trading I getting exception from time to time something about “algorithm spent more than 10 minutes and hence terminated”. How to implement the strategy described above without getting into this issue?

class RollingWindowAlgorithm(QCAlgorithm):
def initialize(self) -> None:
self.set_start_date(2021, 1, 1)
self.set_cash(100000)
self.universe_settings.resolution = Resolution.DAILY
self.add_universe(self.coarse_filter)
# Dictionary to hold rolling windows
self.rolling_windows: dict[Symbol, RollingWindow[TradeBar]] = {}
self.window_size = 100
def coarse_filter(self, coarse) -> list:
# Keep invested securities and filter for US stocks for rolling window
return [x.Symbol for x in coarse if x.Symbol.value.endswith('.O') or self.portfolio[x.Symbol].invested]
def update_rolling_window(self, symbol, price) -> None:
if symbol not in self.rolling_windows:
self.rolling_windows[symbol] = RollingWindow[TradeBar](self.window_size)
self.rolling_windows[symbol].add(TradeBar(self.Time, symbol, price, price, price, price, 100))
def on_data(self, slice) -> None:
for symbol, trade_bar in slice.bars.items():
self.update_rolling_window(symbol, trade_bar.close)
if len(self.rolling_windows) > 0:
# Calculate price change for the last 100 days
changes = {
symbol: (window[0].close - window[self.window_size - 1].close) / window[self.window_size - 1].close
for symbol, window in self.rolling_windows.items() if window.is_ready}
# Sort and take top 10
top_10 = sorted(changes, key=lambda x: changes[x], reverse=True)[:10]
self.rebalance(top_10)
def rebalance(self, top_10) -> None:
for symbol in list(self.active_symbols):
if symbol not in top_10:
self.liquidate(symbol)
for symbol in top_10:
self.set_holdings(symbol, 0.1) # Equal weight
def on_securities_changed(self, changes) -> None:
self.active_symbols = {sec.Symbol for sec in changes.added_securities}

+ Expand

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.

Mia Alissi INVESTOR

4 days ago

class RollingWindowAlgorithm(QCAlgorithm):
    def initialize(self) -> None:
        self.set_start_date(2021, 1, 1)
        self.set_cash(100000)
        self.universe_settings.resolution = Resolution.DAILY
        self.add_universe(self.coarse_filter)
        # Dictionary to hold rolling windows
        self.rolling_windows: dict[Symbol, RollingWindow[TradeBar]] = {}
        self.window_size = 100
    def coarse_filter(self, coarse) -> list:
        # Keep invested securities and filter for US stocks for rolling window
        return [x.Symbol for x in coarse if x.Symbol.value.endswith('.O') or self.portfolio[x.Symbol].invested]
    def update_rolling_window(self, symbol, price) -> None:
        if symbol not in self.rolling_windows:
            self.rolling_windows[symbol] = RollingWindow[TradeBar](self.window_size)
        self.rolling_windows[symbol].add(TradeBar(self.Time, symbol, price, price, price, price, 100))
    def on_data(self, slice) -> None:
        for symbol, trade_bar in slice.bars.items():
            self.update_rolling_window(symbol, trade_bar.close)
        if len(self.rolling_windows) > 0:
            # Calculate price change for the last 100 days
            changes = {
                symbol: (window[0].close - window[self.window_size - 1].close) / window[self.window_size - 1].close
                for symbol, window in self.rolling_windows.items() if window.is_ready}
            # Sort and take top 10
            top_10 = sorted(changes, key=lambda x: changes[x], reverse=True)[:10]
            self.rebalance(top_10)
    def rebalance(self, top_10) -> None:
        for symbol in list(self.active_symbols):
            if symbol not in top_10:
                self.liquidate(symbol)
        for symbol in top_10:
            self.set_holdings(symbol, 0.1)  # Equal weight
    def on_securities_changed(self, changes) -> None:
        self.active_symbols = {sec.Symbol for sec in changes.added_securities}
+ Expand
- Collapse

This corrected version adheres to Python's type hinting and fixes potential attribute errors. Make sure to verify the specific usage of RollingWindow and TradeBar within your QuantConnect environment as appropriate for your implementation.

Sources: Short Term Reversal Strategy In Stocks, Coarse Fundamental Top3 Algorithm, Momentum And State Of Market Filters, Coarse Fine Fundamental Combo Algorithm, Fundamental Factor Long Short Strategy

Upvote

Platform

Radically Open-Source Algorithmic Trading Engine

Join Our Discord Channel

Quarterly Open-Source Trading Competition

Draft Discussions

Bookmarked Discussions

SEARCH DISCUSSIONS

364,800 Quants.

VOTE FOR UPCOMING FEATURES

JOIN OUR Community MAILING LIST

IN THIS RESEARCH

PARTICIPANTS

Actions

Join QuantConnect for Free