Feature construction from market data for ML trading models including price, volume, on-chain, and microstructure features

Feature Engineering for Trading ML Feature engineering is the single highest leverage activity in building ML trading models. Model selection (XGBoost vs. neural net vs. logistic regression) matters far less than the quality and diversity of input features. A simple model on great features will outperform a complex model on raw prices every time. This skill covers constructing, validating, and selecting features from market data for use in classification (signal classification) and regression models targeting crypto/Solana token trading. Why Features Beat Models Raw OHLCV data is non stationary, noisy, and high dimensional. Models trained directly on price series will overfit. Feature engineering transforms raw data into stationary, informative signals that capture distinct aspects of market behavior: Compression : Reduce thousands of price bars to dozens of descriptive statistics Stationarity : Convert non stationary prices into stationary returns and ratios Domain knowledge : Encode trader intuition (support/resistance, volume climax) as computable quantities Regime awareness : Features that behave differently in trending vs. ranging markets help models adapt Feature Categories 1. Price Features Derived purely from OHLCV price columns. These capture trend, momentum, and volatility from the price series itself. Feature Formula Lookback log return ln(close t / close {t 1}) 1 bar abs return abs(log return) 1 bar return volatility std(log return, N) 20 bars momentum N close t / close {t N} 1 5, 10, 20 acceleration momentum 5 momentum 5[5] 10 bars high low range (high low) / close 1 bar close position (close low) / (high low) 1 bar gap open t / close {t 1} 1 1 bar rolling skew skew(log return, N) 20 bars rolling kurtosis kurtosis(log return, N) 20 bars 2. Volume Features Volume confirms or contradicts price movements. Divergences between price and volume are among the most reliable signals in short term trading. Feature Formula Lookback volume ratio volume t / mean(volume, N) 20 bars volume ma ratio sma(volume, 5) / sma(volume, 20) 20 bars obv slope slope(OBV, N) 10 bars vwap deviation (close VWAP) / VWAP intraday volume acceleration volume ratio t volume ratio {t 1} 21 bars buy volume ratio buy volume / total volume 1 bar dollar volume close volume 1 bar volume cv std(volume, N) / mean(volume, N) 20 bars 3. Technical Features Standard technical indicators computed via pandas ta . Use the pandas ta skill for full parameter documentation. Feature Source Lookback rsi RSI(14) 14 bars macd histogram MACD(12,26,9) histogram 33 bars bb position (close BB lower) / (BB upper BB lower) 20 bars bb width (BB upper BB lower) / BB mid 20 bars atr ratio ATR(14) / close 14 bars adx ADX(14) 14 bars stoch k Stochastic %K(14,3) 14 bars cci CCI(20) 20 bars mfi MFI(14) 14 bars supertrend direction Supertrend direction (+1/ 1) 10 bars 4. Microstructure Features Derived from trade level data (individual swaps/transactions). Require on chain or DEX API data. Feature Description trade count ratio Trades this bar / avg trades per bar avg trade size Mean trade size in USD large trade pct % of volume from trades $10k unique traders Count of distinct wallet addresses buy count ratio Buy trades / total trades trade size entropy Shannon entropy of trade size distribution 5. On Chain Features Derived from blockchain state changes. Require Helius or Solana RPC data. Feature Description holder count change Change in unique holders over N periods whale net flow Net tokens moved by top 10 holders token velocity Transfer volume / circulating supply liquidity change Change in DEX liquidity pool TVL 6. Cross Asset Features Capture relationships between the target token and broader market. Feature Description sol correlation Rolling correlation with SOL price btc beta Rolling beta to BTC returns sector momentum Average return of tokens in same sector 7. Time Features Cyclical encoding of calendar time. Use sin/cos encoding to preserve cyclical continuity (hour 23 is close to hour 0). Stationarity Non stationary features will cause your model to fail on new data. A feature is stationary if its statistical properties (mean, variance) don't change over time. Testing for Stationarity Use the Augmented Dickey Fuller (ADF) test: Making Features Stationary Non Stationary Stationary Transform Price Log return Volume Volume ratio (vol / avg vol) OBV OBV slope (regression coefficient) Holder count Holder count change RSI Already stationary (bounded 0 100) Dollar volume Dollar volume / rolling mean Rule : If a feature trends upward or downward over time, it is non stationary. Transform it into a ratio, difference, or rate of change. Normalization After computing features, normalize them so that all features have comparable scales. This is critical for distance based models (KNN, SVM) and helpful for tree models. Method Formula When to Use Z score (x mean) / std Gaussian like distributions Min max (x min) / (max min) Bounded features (RSI, BB position) Rank rank(x) / len(x) Heavy tailed distributions Critical : Use rolling statistics for normalization. Never use full sample mean/std — that introduces lookahead bias. No Lookahead Guarantee The most dangerous bug in trading ML is lookahead bias — using future information to compute features or targets. Follow these rules absolutely: 1. Rolling calculations only : Never use .mean() or .std() on the full series. Always use .rolling(N).mean() . 2. Shift targets forward, not features backward : The target is close.shift( N) / close 1 (future return), not close / close.shift(N) 1 (past return used as target). 3. No future index alignment : When joining feature and target DataFrames, verify that feature row t is paired with target row t (where target already contains the forward shift). 4. Train/test split by time : Never random split. Always train = data[:split idx] , test = data[split idx:] . Feature Selection After computing many features, select the most predictive and least redundant: Step 1: Remove Low Variance Features Step 2: Correlation Filter Remove features with 0.9 correlation to another feature (keep the one with higher target correlation): Step 3: Feature Importance Train a random forest and rank by importance: Step 4: Mutual Information Non linear alternative to correlation: Label Creation Labels (targets) define what the model learns to predict. Binary Classification Typical thresholds: 1% for 1h bars, 3% for 4h bars, 5% for daily bars. Multi Class Classification Regression Binary classification is recommended for initial models — it's simpler and more robust to noise. Integration with Other Skills pandas ta : Compute technical indicators that become features birdeye api : Fetch OHLCV and trade data for feature computation helius api : Fetch on chain data for holder/whale features signal classification : Use engineered features as model inputs regime detection : Regime labels as features or for regime conditional models ohlcv processing : Clean and resample raw data before feature computation Files References references/feature catalog.md — Complete catalog of ~40 features with formulas, lookbacks, stationarity status, and interpretation notes references/pitfalls.md — Common mistakes in trading feature engineering: lookahead bias, overfitting, survivorship bias, data snooping, non stationarity Scripts scripts/build features.py — Compute 25+ features from OHLCV data with stationarity testing and quality reporting. Supports demo mode with synthetic data or live data via Birdeye API. scripts/feature importance.py — Rank features by predictive power using tree based importance and permutation importance. Identifies redundant features via correlation analysis.

feature-engineering