TF-IDF basket · Engle-Granger pairs · Polymarket
Related to Will Google have the best AI model at the end of June 2026?
Markets in the same TF-IDF topic cluster, with a 7-day Engle-Granger cointegration screen against each peer. Pairs with t-statistic below −3.34 reject the null of no cointegration at the 5% level — meaning the two markets share a stationary spread useful for relative-value trades.
§1 · Topic basket · “ai · tech · business” (9 markets)
Top terms: aitechbusinessbestmodelend
- Will Anthropic have the best AI model at the end of June 2026?94.7¢ YES · $56.2k 24h
- ★ Will Google have the best AI model at the end of June 2026?3.8¢ YES · $46.2k 24h
- Will GPT-5.6 not be released by June 28, 2026?23.1¢ YES · $17.1k 24h
- Will Meta have the best AI model at the end of June 2026?0.1¢ YES · $27.9k 24h
- Will Mistral have the best AI model at the end of July 2026?0.1¢ YES · $20.7k 24h
- Will OpenAI have the best AI model at the end of June 2026?1.9¢ YES · $39.7k 24h
- Will xAI have the best AI model at the end of June 2026?0.1¢ YES · $35.6k 24h
- Will Z.ai have the best AI model at the end of July 2026?0.4¢ YES · $16.8k 24h
- Will Z.ai have the best AI model at the end of June 2026?0.1¢ YES · $28.4k 24h
§2 · Engle-Granger cointegration screen (8 pairs tested)
For each peer, 7d aligned series are regressed; the residuals are tested for a unit root (ADF). Reject H₀ (no cointegration) at 5% when t < −3.34. Cointegrated pairs share a stationary spread — candidates for pairs / basket trades.
| Peer | t-stat | n | Verdict |
|---|---|---|---|
| -5.103 | 168 | ▲ cointegrated · pair candidate | |
| -3.736 | 168 | ▲ cointegrated · pair candidate | |
| -3.286 | 168 | no rejection · independent | |
| -2.221 | 168 | no rejection · independent | |
| -1.986 | 168 | no rejection · independent | |
| -1.821 | 168 | no rejection · independent | |
| -1.629 | 168 | no rejection · independent | |
| -1.555 | 168 | no rejection · independent |
▸ Advanced metrics · M2M bundle
polymarket · will-google-have-the-best-ai-model-at-the-end-of-june-2026 · fresh · feed 8s old24h sparkline · 60 pts▼ —
realized vol (ann.)
9.77%
max drawdown
3.95%
sharpe
—
ulcer index
1.77%
RMS drawdown
pain index
1.52%
mean drawdown
mod. VaR 95%
0.00%
Cornish-Fisher
martin ratio
—
ret / ulcer
CDaR 95%
3.95%
cond. drawdown
gain/pain
0.75
Σgain / Σ|loss|
sterling
—
ret / CDaR
omega (θ=0)
0.75
upside/downside
roll spread
0.5 bps
implied (price-only)
bars used
506
store
spread
—
24h Δ
—
flow lean
—
carry
flat
signalNEUTRALconfidence 20%
Same bundle via M2M API:
/api/m2m/pm-will-google-have-the-best-ai-model-at-the-end-of-june-2026/bundle · venue execution: polymarket →