Add request future universe data examples

03 Writing Algorithms/03 Securities/99 Asset Classes/06 Futures/01 Requesting Data/01 Universes/02 Create Universes.html

@@ -1,5 +1,9 @@
 <p>To add a universe of Future contracts, in the <code class="csharp">Initialize</code><code class="python">initialize</code> method, call the <code class="csharp">AddFuture</code><code class="python">add_future</code> method. This method returns an <code>Future</code> object, which has a <code class="csharp">SetFilter</code><code class="python">set_filter</code> method you can call to filter the set of tradable contract down to just the contracts you want.</p>
 
+<h4>Example: Continous Future Filtering And Rollover</h4>
+<p>The following algorithm shows how to request continuous Futures data of E-Mini S&P 500, with a universe filter to obtain only the contracts expiring within 6 months. To access the data of the continuous Future contract and mapped contract, you need to call the <code class="csharp">Symbol</code></p><code class="python">symbol</code> and <code class="csharp">Mapped</code></p><code class="python">mapped</code> properties of the <code>Future</code> object respectively, while accessing all contracts in the filtered Future chain using <code class="csharp">FuturesChains</code></p><code class="python">future_chains</code> of the <code>Slice</code> object.</p>
+
+<p>To rollover the old mapped contract to the new one, you can refer to the below making use of the <code class="csharp">OnSymbolChangedEvents</code></p><code class="python">on_symbol_changed_events</code> event handler.</p>
 <div class="section-example-container">
     <pre class="csharp">public class BasicFutureAlgorithm : QCAlgorithm
 {

03 Writing Algorithms/03 Securities/99 Asset Classes/06 Futures/01 Requesting Data/01 Universes/99 Examples.html

@@ -0,0 +1,209 @@
+<p>The following examples demonstrate some common practices for requesting Futures universe data.</p>
+
+<h4>Example 1: Rollover</h4>
+<p>Future contracts expire monthly or quarterly in most cases. Hence, if we hold Future position in month or quarter end, we must consider rolling over to the mapped contract. The following algorithm shows how to buy and roll over to the next front month Future contract. We make use of the universe selection filter to select the front month contract and order the next mapped contract during the previous one expires.</p>
+<div class="section-example-container">
+    <pre class="csharp">public class FutureExampleAlgorithm : QCAlgorithm
+{
+    private Symbol _future;
+
+    public override void Initialize()
+    {
+        var future = AddFuture(Futures.Indices.SP500EMini, extendedMarketHours: true);
+        _future = future.Symbol;
+        // We only want to hold position of the front month contract.
+        future.SetFilter((u) =&gt; u.OnlyApplyFilterAtMarketOpen().FrontMonth());
+    }
+
+    public override void OnSecuritiesChanged(SecurityChanges changes)
+    {
+        // Liquidate if expired (or not being the front month contract anymore) and exit universe.
+        foreach (var removed in changes.RemovedSecurities)
+        {
+            Liquidate(removed.Symbol);
+        }
+
+        foreach (var added in changes.AddedSecurities)
+        {
+            // Make sure the newly added contract is an actual mapped tradable contract.
+            if (!added.Symbol.IsCanonical())
+            {
+                // Roll over by ordering the same quantity.
+                // Use limit order since market on open order is not supported on Future and avoid extreme quote filling.
+                LimitOrder(added.Symbol, 1m, Securities[_future].Price);
+            }
+        }
+    }
+}</pre>
+    <pre class="python">class FutureExampleAlgorithm(QCAlgorithm):
+    def initialize(self) -&gt; None:
+        future = self.add_future(Futures.Indices.SP_500_E_MINI, extended_market_hours=True)
+        self._future = future.symbol
+        # We only want to hold position of the front month contract.
+        future.set_filter(lambda u: u.only_apply_filter_at_market_open().front_month())
+
+    def on_securities_changed(self, changes: SecurityChanges) -&gt; None:
+        # Liquidate if expired (or not being the front month contract anymore) and exit universe.
+        for removed in changes.removed_securities:
+            self.liquidate(removed.symbol)
+
+        for added in changes.added_securities:
+            # Make sure the newly added contract is an actual mapped tradable contract.
+            if not added.symbol.is_canonical():
+                # Roll over by ordering the same quantity.
+                # Use limit order since market on open order is not supported on Future and avoid extreme quote filling.
+                self.limit_order(added.symbol, 1, self.securities[self._future].price)</pre>
+</div>
+
+<h4>Example 2: Continuous Future Indicator</h4>
+<p>One of the major applications of <a href='/docs/v2/writing-algorithms/universes/futures#12-Continous-Contracts'>Continuous Future</a> is to obtain smooth price series to feed into indicators. This can ensure the indicator gets the correct price data that is comparable to the current mapped Future contract. In this example, we demonstrate a 252-day <a href="/docs/v2/writing-algorithms/indicators/supported-indicators/exponential-moving-average">Exponential Moving Average indicator</a> update using continuous ES contract data.</p>
+<div class="section-example-container">
+    <pre class="csharp">public class FutureExampleAlgorithm : QCAlgorithm
+{
+    private dynamic _future;
+
+    public override void Initialize()
+    {
+        // Use backward ratio normalization for continuous contract to feed smooth, comparable price series to the indicator.
+        _future = AddFuture(Futures.Indices.SP500EMini,
+            dataNormalizationMode: DataNormalizationMode.BackwardsRatio,
+            extendedMarketHours: true);
+        // We only want to hold position of the front month contract.
+        (_future as Future).SetFilter((u) =&gt; u.OnlyApplyFilterAtMarketOpen().FrontMonth());
+        // Create a 252-day EMA indicator as a trend estimator.
+        _future.ema = EMA(_future.Symbol, 252, Resolution.Daily);
+        // Warm up the EMA indicator to make it readily available.
+        WarmUpIndicator((Symbol)_future.Symbol, (ExponentialMovingAverage)_future.ema);
+    }
+
+    public override void OnData(Slice slice)
+    {
+        // Ensure the TradeBar data is available for the Future. Only use updated price data to update the indicator and make trading decision.
+        if (slice.Bars.ContainsKey(_future.Symbol))
+        {
+            // Buy the mapped contract if the trend is estimated to go up (price above EMA).
+            if (_future.ema.Current.Value &gt;= slice.Bars[_future.Symbol].Close)
+            {
+                SetHoldings(_future.Mapped, 0.1m);
+            }
+            // Short the mapped contract if the trend is estimated to go down (price below EMA).
+            else
+            {
+                SetHoldings(_future.Mapped, -0.1m);
+            }
+        }
+    }
+
+    public override void OnSecuritiesChanged(SecurityChanges changes)
+    {
+        // Liquidate if expired (or not being the front month contract anymore) and exit universe.
+        foreach (var removed in changes.RemovedSecurities)
+        {
+            Liquidate(removed.Symbol);
+        }
+    }
+}</pre>
+    <pre class="python">class FutureExampleAlgorithm(QCAlgorithm):
+    def initialize(self) -&gt; None:
+        # Use backward ratio normalization for continuous contract to feed smooth, comparable price series to the indicator.
+        self._future = self.add_future(Futures.Indices.SP_500_E_MINI,
+            data_normalization_mode=DataNormalizationMode.BACKWARDS_RATIO,
+            extended_market_hours=True)
+        # We only want to hold position of the front month contract.
+        self._future.set_filter(lambda u: u.only_apply_filter_at_market_open().front_month())
+        # Create a 252-day EMA indicator as a trend estimator.
+        self._future.ema = self.ema(self._future.symbol, 252, Resolution.DAILY)
+        # Warm up the EMA indicator to make it readily available.
+        self.warm_up_indicator(self._future.symbol, self._future.ema)
+
+    def on_data(self, slice: Slice) -&gt; None:
+        # Ensure the TradeBar data is available for the Future. Only use updated price data to update the indicator and make trading decision.
+        if self._future.symbol in slice.bars:
+            # Buy the mapped contract if the trend is estimated to go up (price above EMA).
+            if self._future.ema.current.value &gt;= slice.bars[self._future.symbol].close:
+                self.set_holdings(self._future.mapped, 0.1)
+            # Short the mapped contract if the trend is estimated to go down (price below EMA).
+            else:
+                self.set_holdings(self._future.mapped, -0.1)
+
+    def on_securities_changed(self, changes: SecurityChanges) -&gt; None:
+        # Liquidate if expired (or not being the front month contract anymore) and exit universe.
+        for removed in changes.removed_securities:
+            self.liquidate(removed.symbol)</pre>
+</div>
+
+<h4>Example 3: Contango</h4>
+<p>In Future trading, contango refers to the far-to-expiry Future contract price is higher than the spot price due to various reasons, such as storage fee and insurance of the commodities. The following example shows a contango trading by shorting the far contract that the price is above a threshold compared to the front month contract price and buying the front month contract to earn the premium in between.</p>
+<div class="section-example-container">
+    <pre class="csharp">public class FutureExampleAlgorithm : QCAlgorithm
+{
+    private Symbol _future;
+
+    public override void Initialize()
+    {
+        // Allow extended market hours trade, which is common for Future since extended market hour is still popular.
+        var future = AddFuture(Futures.Metals.MicroGold, extendedMarketHours: true);
+        _future = future.Symbol;
+        // Limit the expiration to within 6 months, as the longer the expiration, the higher the price uncertainty.
+        future.SetFilter((u) =&gt; u.OnlyApplyFilterAtMarketOpen().Expiration(0, 183));
+    }
+
+    public override void OnData(Slice slice)
+    {
+        // Get Future chain only for the selected Future contract.
+        if (!Portfolio.Invested && slice.FutureChains.TryGetValue(_future, out var chain))
+        {
+            // It takes 2 contracts with different expiries to form a horizontal spread arbitration to earn price difference in contango.
+            if (chain.Count() &lt; 2) return;
+            var farContract = chain.MaxBy(x =&gt; x.Expiry);
+            var nearContract = chain.MinBy(x =&gt; x.Expiry);
+
+            // Check if the far contract price is 1% higher than the near one.
+            // If so, short the far contract and buy the near one to earn the horizontal spread premium.
+            if (farContract.BidPrice &gt;= nearContract.AskPrice * 1.01m)
+            {
+                MarketOrder(farContract.Symbol, -1);
+                MarketOrder(nearContract.Symbol, 1);
+            }
+        }
+    }
+
+    public override void OnSecuritiesChanged(SecurityChanges changes)
+    {
+        // Liquidate if expired (or not being the front month contract anymore) and exit universe.
+        foreach (var removed in changes.RemovedSecurities)
+        {
+            Liquidate();
+        }
+    }
+}</pre>
+    <pre class="python">class FutureExampleAlgorithm(QCAlgorithm):
+    def initialize(self) -&gt; None:
+        # Allow extended market hours trade, which is common for Future since extended market hour is still popular.
+        future = self.add_future(Futures.Metals.MICRO_GOLD, extended_market_hours=True)
+        self._future = future.symbol
+        # Limit the expiration to within 6 months, as the longer the expiration, the higher the price uncertainty.
+        future.set_filter(lambda u: u.only_apply_filter_at_market_open().expiration(0, 183))
+
+    def on_data(self, slice: Slice) -&gt; None:
+        # Get Future chain only for the selected Future contract.
+        chain = slice.future_chains.get(self._future)
+        if not self.portfolio.invested and chain:
+            # It takes 2 contracts with different expiries to form a horizontal spread arbitration to earn price difference in contango.
+            if len(list(chain)) &lt; 2:
+                return
+            sorted_by_expiry = sorted(chain, key=lambda x: x.expiry)
+            far_contract = sorted_by_expiry[-1]
+            near_contract = sorted_by_expiry[0]
+
+            # Check if the far contract price is 1% higher than the near one.
+            # If so, short the far contract and buy the near one to earn the horizontal spread premium.
+            if far_contract.bid_price &gt;= near_contract.ask_price * 1.01:
+                self.market_order(far_contract.symbol, -1)
+                self.market_order(near_contract.symbol, 1)
+
+    def on_securities_changed(self, changes: SecurityChanges) -&gt; None:
+        # Liquidate if expired (or not being the front month contract anymore) and exit universe.
+        for removed in changes.removed_securities:
+            self.liquidate()</pre>
+</div>