- C1 TOY (EW-101): OOS ret +1.4%, SR 0.07, maxDD -3.7%, boot-p 0.394
- C2 TOM (EW-101): OOS ret +19.8%, SR 0.24, maxDD -18.6%, boot-p 0.213
- C3 TOQ (EW-101): OOS ret +9.8%, SR 0.19, maxDD -16.0%, boot-p 0.274
- C4 EOM-rev (SPY): OOS ret +136.6%, SR 0.88, maxDD -19.8%, boot-p 0.002
Split 2016-01-01, next-open-style daily hold, 10bps per switch.
ALL DONE
EOM-reversal discrimination test (complementary half)
- Months after negative-EOM months: SPY SR 1.52 (n=1030, OOS) � this is what C4 trades.
- Months after positive-EOM months: SPY SR 0.56 � the 50% of months C4 sits out.
- Discrimination is real (1.52 vs 0.56), but as a sleeve C4 nets OOS SR 0.88 vs SPY B&H 0.89 -> no excess over beta after wrapping into a long-only rule. Parked as a timing prior, not a sleeve.
Verdict
- C1 TOY: REJECT (SR 0.07)
- C2 TOM: REJECT (SR 0.24, p=0.213)
- C3 TOQ: REJECT (SR 0.19, p=0.274)
- C4 EOM-reversal: REJECT as a sleeve (SR 0.88 = B&H); discrimination note kept above.