Creating Features
Learn how to create custom features (technical indicators) using MangoLabs' intuitive interface. This guide will walk you through the entire process, from selecting an indicator type to testing it with live chart visualization.
What You'll Learn
- Step-by-step feature creation process
- Understanding indicator parameters
- Naming conventions and best practices
- Testing and previewing features with charts
Step-by-Step Creation Guide
Creating a feature in MangoLabs is a straightforward process. Let's walk through it:
Step 1Navigate to Features
Click on "Features" in the left sidebar to access your feature library. This is where all your technical indicators are stored and managed.
Step 2Click Create Feature
Click the "Create Feature" button in the top-right corner. This will open the feature creation dialog.
Step 3Fill in Feature Details
Feature Name
A unique identifier for your feature (e.g., RSI_14, SMA_50)
Indicator Type
Select the indicator you want to use from the dropdown (RSI, SMA, MACD, etc.)
Parameters
Configure the indicator-specific parameters (period, source, method, etc.)
Step 4Preview (Optional)
Click "Preview" to visualize how your feature looks on a chart with historical data. This helps verify the indicator is configured correctly.
Step 5Save Feature
Click "Create" to save your feature. It will now be available in your feature library for use in strategies.
Understanding Parameters
Different indicators have different parameters. Here are the most common ones:
Period (Length)
Number of candles to look back for calculation.
Example: RSI period of 14 means the last 14 candles are used to calculate the RSI value.
Source
Which price data to use for calculation.
Options: Close (most common), Open, High, Low, or HL2 (average of High/Low)
Method
Calculation method for moving averages.
Options: SMA (Simple), EMA (Exponential), WMA (Weighted)
Standard Deviation
Multiplier for volatility indicators.
Example: Bollinger Bands typically use 2 standard deviations from the moving average.
Common Feature Examples
Here are step-by-step examples for creating the most popular features:
Example 1: RSI (Relative Strength Index)
CONFIGURATION
- Name:
RSI_14 - Type: RSI
- Period: 14
- Source: Close
USE CASE
Identifies overbought (RSI > 70) and oversold (RSI < 30) conditions for mean reversion strategies.
Pro Tip: Period 14 is standard, but try 9 for faster signals or 21 for smoother signals with less noise.
Example 2: Simple Moving Average (SMA)
CONFIGURATION
- Name:
SMA_50 - Type: SMA
- Period: 50
- Source: Close
USE CASE
Identifies trend direction. Price above SMA = uptrend, price below SMA = downtrend. Great for trend-following strategies.
Pro Tip: Create multiple SMAs (20, 50, 200) to identify different trend timeframes and crossover signals.
Example 3: Bollinger Bands
CONFIGURATION
- Name:
BB_20_2 - Type: Bollinger Bands
- Period: 20
- Std Dev: 2
- Source: Close
USE CASE
Measures volatility and identifies overbought/oversold conditions. Price touching lower band suggests potential buy, upper band suggests sell.
Pro Tip: Bollinger Bands output THREE values: upper band, middle band (SMA), and lower band. You can reference each in your strategy.
Example 4: MACD (Moving Average Convergence Divergence)
CONFIGURATION
- Name:
MACD_12_26_9 - Type: MACD
- Fast Period: 12
- Slow Period: 26
- Signal Period: 9
- Source: Close
USE CASE
Detects momentum changes and trend reversals. MACD crossing above signal line = bullish, below = bearish.
Pro Tip: MACD outputs TWO values: MACD line and Signal line. Look for crossovers between them to generate signals.
Naming Conventions
Following a consistent naming convention makes your features easier to identify and manage:
Recommended Format
INDICATOR_PARAMETER1_PARAMETER2
Examples:
RSI_14- RSI with period 14EMA_50- Exponential Moving Average with period 50BB_20_2- Bollinger Bands with period 20, std dev 2MACD_12_26_9- MACD with fast 12, slow 26, signal 9ATR_14- Average True Range with period 14
Why This Matters: When you have 20+ features in your library, clear names like RSI_14 are much easier to find than generic names like my_indicator_v3.
Testing and Preview
Before using a feature in a strategy, it's a good idea to preview it to ensure it's calculating correctly:
Chart Visualization
- After creating a feature, navigate to your feature library
- Click on the feature card to view details
- Click "View on Chart" to see the indicator plotted on historical data
- Select a symbol (e.g., BTCUSDT) and timeframe (e.g., 1h)
- The chart will display price data with your indicator overlaid
What to Look For: Check that the indicator values make sense. For RSI, values should be between 0-100. For moving averages, the line should smooth out price action. If something looks wrong, review your parameters.
Best Practices and Tips
Start Simple
Begin with standard parameters (RSI-14, SMA-50, etc.) before experimenting with custom values. Standard settings are widely tested and proven.
Test Variations
Create multiple versions of the same indicator with different parameters (RSI_9, RSI_14, RSI_21) to find what works best for your strategy.
Document Purpose
Use the description field to note what each feature is for (e.g., "Fast RSI for scalping strategies" or "Long-term trend SMA").
Reuse Features
Before creating a new feature, check if you already have one with the same parameters. Reusing features improves performance through caching.
Common Mistakes to Avoid
- Period Too Short: Very short periods (e.g., RSI-5) can generate excessive false signals and noise.
- Period Too Long: Very long periods (e.g., SMA-200 on 1m timeframe) may lag too much to be useful.
- Duplicate Features: Creating RSI_14 multiple times wastes resources. Check your library first.
- No Testing: Always preview your features on charts before using them in strategies to catch configuration errors.