Market Psychology

After a fall from 60, the last ten days trade in a box 7.3% wide. Inside it, average volume on up days is 1.5 times volume on down days. A Wyckoff reader calls that accumulation: someone buying on rallies while sellers dry up. The code finds the pattern; the story about who is behind it is an interpretation.

• The “Composite Man” is a narrative device. The same volume can be read as accumulation or distribution, and no peer-reviewed test has validated Wyckoff’s schematics.
• Volume does carry information about prices (Blume, Easley & O’Hara 1994), which is the part worth keeping.
• A range can resolve either way; the label is only confirmed by the breakout.

df = pd.DataFrame({                              # a fall, then a sideways range, made up
    'close':  [60, 56, 52, 49, 47, 48, 46, 47.5, 47, 48.5, 47.5, 49, 48, 49.5],
    'volume': [30, 35, 40, 45, 50, 38, 22, 36, 20, 40, 21, 42, 19, 44]})
box = df.tail(10)                                    # the last ten days
width = (box.close.max() - box.close.min()) / box.close.mean()
up = box.close.diff() > 0
up_down_volume = box.volume[up].mean() / box.volume[~up].mean()   # > 1 is read as quiet buying

Assumes import numpy as np and import pandas as pd. Each snippet carries its own example numbers; the comments say which are made up.