Helpful Methods & Indicators

Here’s a collection of essential Pandas functions, tips, tricks, and indicators commonly used in trading metric and signal calculations. Even when not indicators themselves, many of these form the foundation for financial indicators that power smart buy/sell decisions.

You’ll notice a bunch of the following indicators/metrics make use of the rolling() function. Read more about it here. The ewm() method is also integral to many trading indicators. You can read more about it here.

Basic Utilities and Metrics

Simple Moving Average (SMA)

data['sma_10'] = data['close'].rolling(window=10).mean()

Calculates the average closing price over the last 10 periods to smooth out short-term fluctuations.

Exponential Moving Average (EMA)

data['ema_10'] = data['close'].ewm(span=10, adjust=False).mean()

Applies more weight to recent prices, making it more responsive to price changes than SMA.

Average Gain

data['gain'] = data['close'].diff().clip(lower=0).rolling(window=14).mean()

Measures the average positive price changes over a 14-period window.

Average Loss

data['loss'] = -data['close'].diff().clip(upper=0).rolling(window=14).mean()

Measures the average negative price changes over a 14-period window.

Relative Strength Index (RSI)

rs = data['gain'] / data['loss']
data['rsi_14'] = 100 - (100 / (1 + rs))

A momentum indicator that measures the speed and change of price movements, ranging from 0 to 100.

Rate of Change (ROC)

data['roc_10'] = data['close'].pct_change(periods=10) * 100

Measures the percentage change in price over the last 10 periods to identify momentum shifts.

Rolling Maximum (Highest Close Over Previous N Days)

data['rolling_max_20'] = data['close'].shift().rolling(window=20).max()

Finds the highest closing price over the last 20 periods to identify resistance levels.

Rolling Minimum (Lowest Close Over Previous N Days)

data['rolling_min_20'] = data['close'].shift().rolling(window=20).min()

Finds the lowest closing price over the last 20 periods to identify support levels.

Indicators

Why not take it one step farther while we’re at it, and put our Pandas knowledge to the test?

Commodity Channel Index (CCI)

data['tp'] = (data['high'] + data['low'] + data['close']) / 3
data['sma_tp'] = data['tp'].rolling(window=20).mean()
data['mad'] = data['tp'].rolling(window=20).apply(lambda x: pd.Series(x).mad())
data['cci_20'] = (data['tp'] - data['sma_tp']) / (0.015 * data['mad'])

Measures how far the price is from its average to detect trends.

Williams %R (Momentum Indicator)

data['highest_high'] = data['high'].rolling(window=14).max()
data['lowest_low'] = data['low'].rolling(window=14).min()
data['williams_r_14'] = -100 * (data['highest_high'] - data['close']) / (data['highest_high'] - data['lowest_low'])

Shows the current price relative to the highest high over 14 periods.

Ulcer Index (Risk Indicator)

data['drawdown'] = (data['close'] - data['close'].cummax()) / data['close'].cummax()
data['ulcer_index'] = data['drawdown'].rolling(window=14).apply(lambda x: (x**2).mean()**0.5)

Tracks downside risk by measuring the depth and duration of drawdowns.

Chande Momentum Oscillator (CMO)

data['cmo_numerator'] = data['gain'].rolling(window=14).sum() - data['loss'].rolling(window=14).sum()
data['cmo_denominator'] = data['gain'].rolling(window=14).sum() + data['loss'].rolling(window=14).sum()
data['cmo_14'] = 100 * (data['cmo_numerator'] / (data['cmo_denominator'] + 1e-10))

Measures trend strength by comparing the sum of gains and losses.

Donchian Channels (Breakout Indicator)

data['donchian_high_20'] = data['high'].shift()rolling(window=20).max()
data['donchian_low_20'] = data['low'].shift()rolling(window=20).min()

Shows the highest and lowest prices over a set period to detect breakouts.

Detrended Price Oscillator (DPO)

data['sma_20'] = data['close'].rolling(window=20).mean()
data['dpo'] = data['close'].shift(10) - data['sma_20']

Removes long-term trends to focus on shorter-term cycles.

Chaikin Money Flow (CMF - Volume-Based Indicator)

data['mfm'] = ((data['close'] - data['low']) - (data['high'] - data['close'])) / (data['high'] - data['low'] + 1e-10)
data['mfv'] = data['mfm'] * data['volume']
data['cmf_20'] = data['mfv'].rolling(window=20).sum() / data['volume'].rolling(window=20).sum()

Analyzes buying and selling pressure based on price and volume movement.

Z-Score (Standardized Price Deviation)

data['rolling_mean'] = data['close'].rolling(window=20).mean()
data['rolling_std'] = data['close'].rolling(window=20).std()
data['z_score'] = (data['close'] - data['rolling_mean']) / (data['rolling_std'] + 1e-10)

Measures price deviations relative to its mean for spotting outliers.