Reversed Halloween (long May-Oct)

A deliberately meaningless test — reversed Halloween (long May-Oct) — run beside the real ones to prove the screen can come back empty.

Conclusion: Placebo

A deliberately meaningless signal, run alongside the real ones. It found nothing, which is the result it was included to produce — had it scored, the whole screen would be suspect.

Every test, against the significance threshold

persistence-robust p · left is stronger evidence
00.050.250.511m
Filled markers fall below the 0.05 threshold before multiple-testing correction; hollow markers do not. Correction is applied across the whole batch, so an individual marker clearing the line is not a result on its own.
Best p
0.204
directional
Best q
BH-FDR
Fired
0/1
at p<0.05
Survived
0
q ≤ 0.10

Method

Hypothesis

Reversed Halloween (long May-Oct)

Rule

prior = +0 # none: placebo
function test(horizon h):
basket = inverse_vol_weighted_mean(instruments)
y = sum(basket, t+1 .. t+h) # strictly forward
x = signal[t] # known at t
beta, t_stat = OLS(y ~ x) with Newey-West HAC
null = [slope(y, roll(x, k)) for k in all offsets]
p = (count(null beyond beta) + 1) / (len(null) + 1)
return beta, t_stat, p
reject if Benjamini-Hochberg q <= 0.10 across the whole screen

Evaluation

Table 1 · Test results

1m+2.5-2.5
HorizonnβHAC tHAC pShift p (dir)Shift p (2-sided)
1m658-0.0073-2.460.0140.2040.204
Directional p-value scores the pre-declared prior; the two-sided value makes no directional assumption.

Reference

Instruments

last price and one-month change, live from Yahoo Finance
SymbolNotes
^GSPCS&P 500 — market index, in USD.
^N225Nikkei 225 — market index, in JPY.
^AXJOS&P/ASX 200 — market index, in AUD.
^BVSPIBOVESPA — market index, in BRL.
^NSEINIFTY 50 — market index, in INR.
^KLSEFTSE Bursa Malaysia KLCI — market index, in MYR.
^NZ50S&P/NZX 50 Gross — market index, in NZD.

Sample

monthly, per signal availability

Sources

NASA Goddard Institute for Space Studies. GISS Surface Temperature Analysis (GISTEMP). v4, global land–ocean temperature anomaly. data.giss.nasa.gov/gistemp
NOAA Climate Prediction Center. Teleconnection indices. AO, NAO, PNA, SAM. www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/ao.shtml
NOAA Physical Sciences Laboratory. Climate indices. ONI, Niño 3.4, MEI v2, DMI, PDO, AMO, TNA, QBO, solar flux. psl.noaa.gov/data/climateindices/list
Royal Observatory of Belgium. Sunspot Number. SILSO, monthly mean. www.sidc.be/SILSO/datafiles
Yahoo Finance. Historical market prices. Adjusted close, month-end. Retrieved with yfinance. finance.yahoo.com

Revision history

RevStageStatusChange
01DraftDirection and criteria fixed before the experiment ran
02TestedPlacebo1 tests at 1 and 3 months; corrected across the 94-test screen
algorithmic-trading-anthology-july-2026-reversed-halloween-placebo | d/rksci