Learn · Tests

Test H7 · 2016–2023

Amavasya (vs Purnima) Beat 17% of shifted calendars

Claim: Volatile stocks do worse.

How it was tested

Window: Two trading days either side of each Amavasya (new moon), compared with two days either side of each Purnima (full moon). Stocks: the Volatile third of liquid NSE stocks (highest one-year volatility), which astrology assigns to the fast planets; groups are re-formed each month using data up to two days earlier.

Measure: the group's average daily return on window days minus all other days. Control: the same measure on 2,000 placebo calendars, the real sky shifted by random amounts, which keeps each window's length and rhythm. The claim passes only if the real sky beats at least 95% of them and still clears the bar after the Holm correction for all tests.

Result

Development data 2016–2023; the 2024–26 data stays sealed until it is opened once. Effects are averages over the period, before any trading costs unless stated. How we test · all results