A/D Line in Python — Coding It from Scratch (pandas)

The A/D Line is the CMF money flow volume, accumulated instead of windowed — a two-line indicator in pandas. The explainer covers the concept; here’s the code, plus the Chaikin Oscillator bonus.

What you’ll need

• Python 3.9+, pandas/numpy
• high, low, close, volume series.

From scratch

import numpy as np
import pandas as pd


def ad_line(high, low, close, volume):
    rng = (high - low).replace(0, np.nan)              # guard flat bars
    mfm = ((close - low) - (high - close)) / rng       # money flow multiplier
    mfm = mfm.fillna(0.0)
    mfv = mfm * volume                                 # money flow volume
    return mfv.cumsum().rename("ad")                   # running total

The multiplier is the close-location value; multiplying by volume signs and scales it; cumsum() accumulates it into the running A/D Line from the output chart. The only difference from CMF is cumsum() instead of a windowed, normalized sum.

The Chaikin Oscillator

Chaikin’s companion is a MACD of the A/D Line:

ad = ad_line(df["high"], df["low"], df["close"], df["volume"])
chaikin_osc = (ad.ewm(span=3, adjust=False).mean()
               - ad.ewm(span=10, adjust=False).mean())

That turns the cumulative line into a zero-centered momentum reading of money flow — a handy, boundedish signal derived from the same data.

Divergence, in code

win = 20
price_up = df["close"] > df["close"].shift(win)
ad_down = ad < ad.shift(win)
df["bearish_div"] = price_up & ad_down    # price up, money flow down

bearish_div flags where price rose but the A/D Line fell — the distribution warning to test in AlgoGen.

Gotchas

• Flat-bar guard. high == low divides by zero in the multiplier; replace/fillna set those bars neutral.
• Cumulative, not windowed. cumsum() is what makes it the A/D Line; window-and- normalize instead and you’ve built CMF.
• Read slope, not level. The absolute value is arbitrary; watch direction and divergence.

The lazy one-liner

import pandas_ta as ta
df["ad"] = ta.ad(df["high"], df["low"], df["close"], df["volume"])

Same indicator elsewhere: MQL5, Pine Script, EasyLanguage, NinjaScript.

---

This post is educational, not financial advice. Indicators describe the past; they don’t predict the future. Backtest anything before you risk real money on it.