ExtractAlpha

Tactical

Introduction

The Tactical dataset by ExtractAlpha is a stock scoring algorithm that captures the technical dynamics of individual US Equities over one to ten trading day horizons. It can assist a longer-horizon investor in timing their entry or exit points or be used in combination with existing systematic or qualitative strategies with similar holding periods.

The data covers a dynamic universe of around 4,700 US Equities per day on average, starts in January 2000, and is delivered on a daily frequency. The Tactical dataset expands upon simple reversal, liquidity, and seasonality factors to identify stocks that are likely to trend or reverse.

This dataset depends on the US Equity Security Master dataset because the US Equity Security Master dataset contains information on splits, dividends, and symbol changes.

For more information about the Tactical dataset, including CLI commands and pricing, see the dataset listing.

About the Provider

ExtractAlpha was founded by Vinesh Jha in 2013 with the goal of providing alternative data for investors. ExtractAlpha's rigorously researched data sets and quantitative stock selection models leverage unique sources and analytical techniques, allowing users to gain an investment edge.

Getting Started

The following snippet demonstrates how to request data from the Tactical dataset:

self.aapl = self.add_equity("AAPL", Resolution.DAILY).symbol
self.dataset_symbol = self.add_data(ExtractAlphaTacticalModel, self.aapl).symbol
_symbol = AddEquity("AAPL", Resolution.Daily).Symbol;
_datasetSymbol = AddData<ExtractAlphaTacticalModel>(_symbol).Symbol;

Data Summary

The following table describes the dataset properties:

PropertyValue
Start DateJanuary 2000
Asset Coverage5,000 US Equities
Data DensitySparse
ResolutionDaily
TimezoneUTC

Requesting Data

To add Tactical data to your algorithm, call the AddDataadd_data method. Save a reference to the dataset Symbol so you can access the data later in your algorithm.

class ExtractAlphaTacticalModelDataAlgorithm(QCAlgorithm):
    def initialize(self) -> None:
        self.set_start_date(2019, 1, 1)
        self.set_end_date(2020, 6, 1)
        self.set_cash(100000)

        self.aapl = self.add_equity("AAPL", Resolution.DAILY).symbol
        self.dataset_symbol = self.add_data(ExtractAlphaTacticalModel, self.aapl).symbol
public class ExtractAlphaTacticalModelDataAlgorithm : QCAlgorithm
{
    private Symbol _symbol, _datasetSymbol;

    public override void Initialize()
    {
        SetStartDate(2019, 1, 1);
        SetEndDate(2020, 6, 1);
        SetCash(100000);

        _symbol = AddEquity("AAPL", Resolution.Daily).Symbol;
        _datasetSymbol = AddData<ExtractAlphaTacticalModel>(_symbol).Symbol;
    }
}

Accessing Data

To get the current Tactical data, index the current Slice with the dataset Symbol. Slice objects deliver unique events to your algorithm as they happen, but the Slice may not contain data for your dataset at every time step. To avoid issues, check if the Slice contains the data you want before you index it.

def on_data(self, slice: Slice) -> None:
    if slice.contains_key(self.dataset_symbol):
        data_point = slice[self.dataset_symbol]
        self.log(f"{self.dataset_symbol} score at {slice.time}: {data_point.score}")
public override void OnData(Slice slice)
{
    if (slice.ContainsKey(_datasetSymbol))
    {
        var dataPoint = slice[_datasetSymbol];
        Log($"{_datasetSymbol} score at {slice.Time}: {dataPoint.Score}");
    }
}

To iterate through all of the dataset objects in the current Slice, call the Getget method.

def on_data(self, slice: Slice) -> None:
    for dataset_symbol, data_point in slice.get(ExtractAlphaTacticalModel).items():
        self.log(f"{dataset_symbol} score at {slice.time}: {data_point.score}")
public override void OnData(Slice slice)
{
    foreach (var kvp in slice.Get<ExtractAlphaTacticalModel>())
    {
        var datasetSymbol = kvp.Key;
        var dataPoint = kvp.Value;
        Log($"{datasetSymbol} score at {slice.Time}: {dataPoint.Score}");
    }
}

Historical Data

To get historical Tactical data, call the Historyhistory method with the dataset Symbol. If there is no data in the period you request, the history result is empty.

# DataFrame
history_df = self.history(self.dataset_symbol, 100, Resolution.DAILY)

# Dataset objects
history_df = self.history[ExtractAlphaTacticalModel](self.dataset_symbol, 100, Resolution.DAILY)
var history = History<ExtractAlphaTacticalModel>(_datasetSymbol, 100, Resolution.Daily);

For more information about historical data, see History Requests.

Remove Subscriptions

To remove a subscription, call the RemoveSecurityremove_security method.

self.remove_security(self.dataset_symbol)
RemoveSecurity(_datasetSymbol);

If you subscribe to Tactical data for assets in a dynamic universe, remove the dataset subscription when the asset leaves your universe. To view a common design pattern, see Track Security Changes.

Example Applications

The Tactical dataset enables you to gain insight into short-term stock dynamics for trading. Examples include the following strategies:

  • Optimizing entry and exit times in a portfolio construction model.
  • Using the raw factor values as technical indicators.
  • Inputting the data into machine learning classifier models as trend/reversal labels.

Classic Algorithm Example

The following example algorithm creates a dynamic universe of the 100 most liquid US Equities. Each day, it then forms an equal-weighted dollar-neutral portfolio of the 10 companies that are most likely to outperform and the 10 that are most likely to underperform.

from AlgorithmImports import *

class ExtractAlphaTacticalModelAlgorithm(QCAlgorithm):

    def initialize(self) -> None:
        self.set_start_date(2024, 9, 1)
        self.set_end_date(2024, 12, 31)
        self.set_cash(100_000)
        self.add_universe(self._select_assets)

    def _select_assets(self, coarse: List[Fundamental]) -> List[Symbol]:
        # Select non-penny stocks with highest dollar volume due to better informed information from more market activities
        # Only the ones with fundamental data are supported by tactical data
        sorted_by_dollar_volume = sorted(
            [x for x in coarse if x.has_fundamental_data and x.price > 4], 
            key=lambda x: x.dollar_volume
        )
        return [x.symbol for x in sorted_by_dollar_volume[-100:]]

    def on_data(self, slice: Slice) -> None:
        # Get the current data from the tactical dataset.
        points = slice.get(ExtractAlphaTacticalModel)
        ## Demonstrate how to iterate through the data and access its members:
        #for dataset_symbol, model in points.items():
        #    self.quit(
        #        f"{self.time} -- "
        #        f"Asset Symbol: {dataset_symbol.underlying}; " 
        #        f"Score: {model.score} "
        #    )

        # Remove assets that have no price.
        points = [x for x in points.items() if self.securities[x[0].underlying].price]
        # Only rebalance when there are new tactical factors.
        if not points:
            return

        # Long the ones with the highest return estimates riding from tactical strategies
        # Short the lowest that predicted stock price goes down
        sorted_by_score = sorted(points, key=lambda x: x[1].score)
        long_symbols = [x[0].underlying for x in sorted_by_score[-10:]]
        short_symbols = [x[0].underlying for x in sorted_by_score[:10]]
        long_targets = [PortfolioTarget(symbol, 0.05) for symbol in long_symbols]
        short_targets = [PortfolioTarget(symbol, -0.05) for symbol in short_symbols]
        self.set_holdings(long_targets + short_targets, True)

    def on_securities_changed(self, changes: SecurityChanges) -> None:
        for security in changes.added_securities:
            # Requesting tactical data for trading signal generation
            security.tactical_model = self.add_data(
                ExtractAlphaTacticalModel, security.symbol
            ).symbol
            # Historical Data
            history = self.history(security.tactical_model, 2, Resolution.DAILY)
        for security in changes.removed_securities:
            # Remove the tactical model data for this asset when it leaves the universe.
            self.remove_security(security.tactical_model)
        
public class ExtractAlphaTacticalModelAlgorithm : QCAlgorithm
{    
    public override void Initialize()
    {
        SetStartDate(2024, 9, 1);
        SetEndDate(2024, 12, 31);
        SetCash(100000);
        AddUniverse(SelectAssets);
    }
    
    private IEnumerable<Symbol> SelectAssets(IEnumerable<Fundamental> coarse)
    {
        // Select non-penny stocks with highest dollar volume due to better informed information from more market activities
        // Only the ones with fundamental data are supported by tactical data
        return (from c in coarse
                where c.HasFundamentalData && c.Price > 4
                orderby c.DollarVolume descending
                select c.Symbol).Take(100);
    }
    
    public override void OnData(Slice slice)
    {
        // Get the current data from the tactical model.
        var points = slice.Get<ExtractAlphaTacticalModel>()
            // Drop factors for assets that have no price.
            .Where(kvp => Securities[kvp.Key.Underlying].Price != 0);
        //// Demonstrate how to iterate through the data and access its members:
        //foreach(var kvp in points)
        //{
        //    var datasetSymbol = kvp.Key;
        //    var model = kvp.Value;
        //    Quit(
        //        $"{Time} -- " +
        //        $"Asset Symbol: {datasetSymbol.Underlying}; " +
        //        $"Score: {model.Score}"
        //    );
        //}

        // Only rebalance when there are new tactical factors.
        if (points.Count() == 0)
        {
            return;
        }

        // Long the ones with the highest return estimates riding from tactical strategies
        // Short the lowest that predicted stock price goes down
        var sortedByScore =  points.OrderBy(kvp => kvp.Value.Score)
            .Select(kvp => kvp.Key.Underlying);
        var longSymbols = sortedByScore.TakeLast(10);
        var shortSymbols = sortedByScore.Take(10);

        // Invest equally and dollar-neutral to evenly dissipate capital risk and hedge systematic risk
        var targets = new List<PortfolioTarget>();
        targets.AddRange(longSymbols.Select(symbol => new PortfolioTarget(symbol, 0.05m)));
        targets.AddRange(shortSymbols.Select(symbol => new PortfolioTarget(symbol, -0.05m)));
        
        SetHoldings(targets, true);
    }
    
    public override void OnSecuritiesChanged(SecurityChanges changes)
    {
        foreach(dynamic security in changes.AddedSecurities)
        {
            // Requesting tactical data for trading signal generation
            security.TacticalModel = AddData<ExtractAlphaTacticalModel>(security.Symbol).Symbol;
            // Historical Data
            var history = History(security.TacticalModel, 2, Resolution.Daily);
        }
        foreach (dynamic security in changes.RemovedSecurities)
        { 
            // Remove the tactical data for this asset when it leaves the universe.
            RemoveSecurity(security.TacticalModel);
        }
    }
}

Framework Algorithm Example

The following example algorithm creates a dynamic universe of the 100 most liquid US Equities. Each day, it then forms an equal-weighted dollar-neutral portfolio of the 10 companies that are most likely to outperform and the 10 that are most likely to underperform.

from AlgorithmImports import *

class ExtractAlphaTacticalModelFrameworkAlgorithm(QCAlgorithm):

    def initialize(self) -> None:
        self.set_start_date(2024, 9, 1)
        self.set_end_date(2024, 12, 31)
        self.set_cash(100_000)
        self.add_universe_selection(LiquidEquitiesUniverseSelectionModel())
        # Custom alpha model to generate trading signal based on tactical data
        self.add_alpha(ExtractAlphaTacticalModelAlphaModel())
        # Invest equally and dollar-neutral to evenly dissipate capital risk and hedge systematic risk
        self.set_portfolio_construction(EqualWeightingPortfolioConstructionModel())
        self.set_execution(ImmediateExecutionModel())


class LiquidEquitiesUniverseSelectionModel(FundamentalUniverseSelectionModel):

    def select(self, algorithm: QCAlgorithm, fundamentals: List[Fundamental]) -> List[Symbol]:
        # Select non-penny stocks with highest dollar volume due to better informed information from more market activities
        # Only the ones with fundamental data are supported by tactical data
        sorted_by_dollar_volume = sorted(
            [x for x in fundamentals if x.has_fundamental_data and x.price > 4], 
            key=lambda x: x.dollar_volume
        )
        return [x.symbol for x in sorted_by_dollar_volume[-100:]]


class ExtractAlphaTacticalModelAlphaModel(AlphaModel):

    def update(self, algorithm: QCAlgorithm, slice: Slice) -> List[Insight]:
        # Get the current data from the tactical dataset.
        points = slice.get(ExtractAlphaTacticalModel)
        ## Demonstrate how to iterate through the data and access its members:
        #for dataset_symbol, model in points.items():
        #    algorithm.quit(
        #        f"{algorithm.time} -- "
        #        f"Asset Symbol: {dataset_symbol.underlying}; " 
        #        f"Score: {model.score} "
        #    )

        # Remove assets that have no price or score.
        points = [x for x in points.items() if algorithm.securities[x[0].underlying].price]

        # Long the ones with the highest return estimates riding from tactical strategies
        # Short the lowest that predicted stock price goes down
        sorted_by_score = sorted(points, key=lambda x: x[1].score)
        insight_directions = (
            [(x, InsightDirection.DOWN) for x in sorted_by_score[:10]] +
            [(x, InsightDirection.UP) for x in sorted_by_score[-10:]]
        )
        return [
            Insight.price(x[0].underlying, Expiry.END_OF_DAY, direction)
            for x, direction in insight_directions
        ]

    def on_securities_changed(self, algorithm: QCAlgorithm, changes: SecurityChanges) -> None:
        for security in changes.added_securities:
            # Requesting tactical data for trading signal generation
            security.tactical_model = algorithm.add_data(
                ExtractAlphaTacticalModel, security.symbol
            ).symbol
            # Historical Data
            history = algorithm.history(security.tactical_model, 2, Resolution.DAILY)
        for security in changes.removed_securities:
            # Remove the tactical model data for this asset when it leaves the universe.
            algorithm.remove_security(security.tactical_model)

public class ExtractAlphaTacticalModelFrameworkAlgorithm : QCAlgorithm
{
    public override void Initialize()
    {
        SetStartDate(2024, 9, 1);
        SetEndDate(2024, 12, 31);
        SetCash(100000);
        AddUniverseSelection(new LiquidEquitiesUniverseSelectionModel());
        // Custom alpha model to generate trading signal based on tactical data
        AddAlpha(new ExtractAlphaTacticalModelAlphaModel());
        // Invest equally and dollar-neutral to evenly dissipate capital risk and hedge systematic risk
        SetPortfolioConstruction(new EqualWeightingPortfolioConstructionModel());
        SetExecution(new ImmediateExecutionModel());
    }
}

public class LiquidEquitiesUniverseSelectionModel : FundamentalUniverseSelectionModel
{
    public override IEnumerable<Symbol> Select(QCAlgorithm algorithm, IEnumerable<Fundamental> fundamentals)
    {
        // Select non-penny stocks with highest dollar volume due to better informed information from more market activities
        // Only the ones with fundamental data are supported by tactical data
        return (from f in fundamentals
                where f.HasFundamentalData && f.Price > 4
                orderby f.DollarVolume descending
                select f.Symbol).Take(100);
    }
}

public class ExtractAlphaTacticalModelAlphaModel: AlphaModel
{    
    public override IEnumerable<Insight> Update(QCAlgorithm algorithm, Slice slice)
    {        
        // Get the current data from the tactical model.
        var points = slice.Get<ExtractAlphaTacticalModel>()
            // Drop factors for assets that have no price or score.
            .Where(kvp => algorithm.Securities[kvp.Key.Underlying].Price != 0);
        //// Demonstrate how to iterate through the data and access its members:
        //foreach(var kvp in points)
        //{
        //    var datasetSymbol = kvp.Key;
        //    var model = kvp.Value;
        //    algorithm.Quit(
        //        $"{algorithm.Time} -- " +
        //        $"Asset Symbol: {datasetSymbol.Underlying}; " +
        //        $"Score: {model.Score}"
        //    );
        //}

        // Long the ones with the highest return estimates riding from tactical strategies
        // Short the lowest that predicted stock price goes down
        var sortedByScore =  points.OrderBy(kvp => kvp.Value.Score)
            .Select(kvp => kvp.Key.Underlying);
        var longSymbols = sortedByScore.TakeLast(10);
        var shortSymbols = sortedByScore.Take(10);

        var insights = new List<Insight>();
        insights.AddRange(longSymbols.Select(symbol => new Insight(symbol, Expiry.EndOfDay, InsightType.Price, InsightDirection.Up)));
        insights.AddRange(shortSymbols.Select(symbol => new Insight(symbol, Expiry.EndOfDay, InsightType.Price, InsightDirection.Down)));
        return insights;
    }

    public override void OnSecuritiesChanged(QCAlgorithm algorithm, SecurityChanges changes)
    {
        foreach(dynamic security in changes.AddedSecurities)
        {
            // Requesting tactical data for trading signal generation
            security.TacticalModel = algorithm.AddData<ExtractAlphaTacticalModel>(security.Symbol).Symbol;
            // Historical Data
            var history = algorithm.History(security.TacticalModel, 2, Resolution.Daily);
        }
        foreach (dynamic security in changes.RemovedSecurities)
        { 
            // Remove the tactical model data for this asset when it leaves the universe.
            algorithm.RemoveSecurity(security.TacticalModel);
        }
    }
}

Data Point Attributes

The Tactical dataset provides ExtractAlphaTacticalModel objects, which have the following attributes:

You can also see our Videos. You can also get in touch with us via Discord.

Did you find this page helpful?

Contribute to the documentation: