How To Use
Learn how easy it is to build automated trading strategies using Build Alpha's point
and click interface. No programming necessary.
This is the screen that allows the user to set up the test/simulation exactly how they wish.
- Long or Short
- Market of Choice
- Test Dates
- Entry Signals
- Exit Criteria
- In Sample Minimum Trades
- Out of Sample Minimum Trades
- Out of Sample Percent
- Fitness Function
Currently offer over 30 futures markets, over 30 ETFs, and over 30 Forex pairs. Almost all have data available for more than 10+ years. Some dating as far back as 1960s. Please contact about specific dates, markets, etc.
- eMini S&P500
- eMini Nasdaq
- eMini Dow Jones
- eMini Russel 2000
- eMini Nikkei
- German Dax Futures
- US 2 Year Futures
- US 5 Year Futures
- US 10 Year Futures
- US 30 Year Futures
- German 10 Year Bund
- Lean Cattle
- Lean Hogs
- Orange Juice
- Natural Gas
- Dollar Index
Forex now available!
**You can also import your own data quite simply – view the buildalpha.com/demo page to see how**
We have an extensive and ever growing list of entry signals. The software will search for the best combinations of these signals based upon the user selected fitness function.
Currently, we offer over 5,000 pre-built entry signals. We include all open, high, low, close combinations over the past 10 bars. We have also included most major technical indicators and oscillators such as: MACD, RSI, Stochastics, ATR, Kaufman Efficiency Ratio, Hurst Exponent, DMI, Moving Averages, Composite Indicators, etc. We have also included candlesticks, volume analysis, consecutive OHLC (open,high,low,close) conditions, seasonality, and much more including secondary data signals such as the VIX.
Build Alpha also allows Inter Market signals. That is, you can now create and improve systems for your primary market by taking into account signals generated from a secondary and tertiary market. For example, create an S&P500 system that improves when Gold has above average volume and US Bonds are below their 200 day moving average. Build Alpha also offers multi-timeframe signals. That is, trade the 30 minute chart only when the daily chart confirms.
Build Alpha also has a custom rule/indicator builder. So if you want to test a custom idea or set of rules you can now create and save thousands of custom signals to run through the software in order to find strategies in combination with the pre-built signal choices.
Finally, Build Alpha has a fully capable python environment. Want to write your own indicators/rules to test? You can now use python, if desired.
All 5,000+ pre-built entry signals can also be used as exit signals. The input interface allows the user to select different exit criteria to create their desired combination. There are over 100,000+ different possible combinations. A few of the exit possibilities are maximum holding time, volatility based profit targets and stop losses, profitable closes, highest high over a lookback period and lowest low over a lookback period. Any of the exit specific rules (such as stop loss) can be combined with any/all of the exit signals (RSI or Price Action, e.g.).
All of these exit combinations can be turned off/on and adjusted. This freedom allows the trader or money manager to find the best strategies to fit his/her personal risk/reward specifications.
These are metrics used to compare and evaluate trading systems. Build Alpha currently offers fifteen unique fitness functions for strategy development.
- Winning Percentage
- Net Profit
- Profit to Drawdown Ratio
- Ratio Win to Loss
- Profit Factor
- Average Trade
- Sharpe Ratio
- Soritino Ratio
- System Quality Number
- CPC Ratio
- Compound Annual Growth Rate
- Correlation Coefficient
- System Quality Measure
In short, the trader or money manager will select one of these fitness functions while setting up the input interface. Build Alpha will create possible strategies trying to optimize this fitness function. For example, the trader or money manager selects Sharpe Ratio then Build Alpha will find strategies with the selected entry signals and exit criteria that have the highest Sharpe Ratio. The same search (entry and exits) can be run again but this time finding strategies that have the higher net profit to drawdown (after changing the fitness function).
Yes, it is very easy to data mine and wind up with a “curve-fit” system. However,Build Alpha has functionality to prevent it before ever beginning. For example, the trader or money manager can select a minimum number of trades both in and out of sample. Furthermore, the trader can adjust the in and out of sample periods by setting the out of sample percent. Setting the out of sample percent for 30% would partition the data into a 70% in sample portion and a 30% out of sample portion.
Theoretically, the user could select 30% out of sample and fail to set a minimum number of trades for out of sample trading. In this example, the trader could find a strategy that rarely traded during the out of sample period due to a regime change or some other factor.
The Build Alpha software allows the trader or money manager the ability to prevent this type of behavior or allow it (as it may be useful to find a strategy that is quiet during a certain regime).
Build Alpha also allows the trader/money manager to view how the strategies generated compare to random strategies created using both random signals and randomly generated data. This helps lower the probabilities that strategies found are “curve-fit” by allowing the user to compare that the strategies found perform significantly better than what could have been found by random signals or random data.
The Output Interface also has measures to prevent data mining bias, curve fitting, and ultimately resulting in the production of seemingly robust trading strategies.
The Output Interface
After selecting the desired exit criteria, market, dates, and entry signals to test the program is set to run. After searching hundreds of thousands of possible strategy combinations the Output Interface will appear.
This interface displays each strategy found, the signals used, its performance criteria such as: profit factor, winning percentage, net p&l, drawdown, number of trades, Sharpe ratio, etc. all in a sortable table.
The Output Interface also allows to view the test results for each period. That is, the trader or money manager can see how a strategy performed in sample, out of sample, and combined.
The Output Interface also allows the visualization of each and every strategy Build Alpha found. The trader money manager can view equity curves(cumulative p&l), edge ratios, Monte Carlo projections, Monte Carlo drawdown analysis, and our own variance testing for how we simulate future trading is likely to proceed for the given strategy.
To see visuals please visit this link: https://buildalpha.com/features/
Additionally, each and every strategy can be added to a portfolio for portfolio evaluation.
Finally, each and every strategy can generate automatable, ready to trade code for TradeStation’s EasyLanguage, MultiChart’s Power Language, and NinjaTrader’s C# code with the click of a button.
Add any strategy into a custom portfolio. View all of our data visualizations on a portfolio, as well. That is, run a Monte Carlo drawdown analysis on a portfolio of strategies to better understand the capital required to trade these with the highest probability for success.
Run our variance test to see where we simulate performance to be n days into the future.
Also, at the click of a button, generate fully automatable/ready to trade TradeStation EasyLanguage, MultiChart PowerLanguage, or NinjaTrader C# code for entire portfolio or combination of strategies.
Inside Portfolio mode the trader or money manager can view how correlated each strategy is with each other. The idea is to construct portfolios of uncorrelated strategies to best optimize the chances of future success.
The correlation matrix uses marked to market daily returns. This allows us to analyze strategies that trade at different frequencies and have different holding periods.
Minimum Variance Portfolio
Minimum Variance Portfolio or Modern Portfolio Theory allows traders and money managers the ability to view the weights to assign each strategy (that can be translated into position sizes) in order to optimize the risk-adjusted returns (Sharpe Ratio) of the portfolio. This unique tool gives traders and money managers an advantage over traditional position sizing approaches.