Mixed — Framler currently rates it neutral / hold based on the multi-factor Framler engine score of 56/100. Top drivers: Defensive / negative MD&A language flagged; Sector in favour — rotation tailwind; Trading near its 52-week high (George-Hwang). Framler combines momentum, quality, value, options flow, insider activity and HMM regime detection to score US and international equities daily. This is statistical research, not investment advice.

Is the read supported by multiple factors, or just one?

The page lists how many of the 13 factors point bullish, bearish, or neutral for CE. Multi-factor alignment indicates stronger conviction; a single-factor signal warrants caution.

Is the uncertainty wide or narrow?

Each ticker shows a 90% Mondrian conformal range (Vovk-Shafer 2005) around the score — the band where the ticker's realised 30-day return rank, against the rest of the universe, is expected to land. It is not a price target. The width comes from the calibration bin, not from how much the factors agree.

Is the read confirmed by sector and regime?

Sector score, BOCPD regime classification (risk-on / transition / risk-off), and the active confluence pattern all appear on the page. Aligned context tightens conviction; misaligned context widens the interval and softens the verdict.

What would invalidate this read?

Factor flips week-over-week, a regime shift, fresh earnings, or insider sells are the main invalidators. Scores update daily; the 30-day score sparkline above the factor breakdown surfaces trend drift in real time.

Basic Materials · NYSE

Calibration 4/4:1d7d30d90d

CE STOCK · Celanese Corporation NeutralMulti-factor aligned

$46.98+1.56%

~1-month outlook, not a day-trade. The score updates daily as the data changes — re-check weekly rather than reacting to a single day's move.

Sector calibration · noisy

Engine accuracy in Basic Materials has been close to zero on recent rolling windows. Treat the composite as informational. Lean on Confluence patterns, Insider clustering, and Options flow — sub-signals that retain edge in narrative-driven cohorts.

AgreementMODERATE ALIGNMENT

Some factor streams agree, others diverge.

90% interval: [9–95]Top driver: NLP Tone

Above-average factor outlook for this regime🔒 Sign in for number

Why CE scores 56

Defensive / negative MD&A language flagged
Sector in favour — rotation tailwind
Trading near its 52-week high (George-Hwang)

Full report — sign in to unlock

🔒 Forward-Return Projection

Expected returns over 1 / 7 / 30 / 90 days with 90% confidence intervals, regime amplifier, pattern multiplier, copula tail-dependence — composed into one Bayesian distribution.

🔒 Full Factor Breakdown

Per-factor 0-100 scores across quality, value, momentum, PEAD, accruals, insider flow, NLP tone, options flow, short interest, sector momentum, factor interaction — plus the score waterfall.

🔒 Confluence Patterns

Active multi-factor patterns (QUALITY COMPOUNDER, VALUE TRAP, SHORT SQUEEZE, OVEREXTENDED, etc.) with academic basis and pattern delta against the raw composite.

🔒 Plain-English Explanation & Scenarios

Plain-English read of what the engine sees on this ticker today + what-if scenarios (price ±N%, regime flip, shock) replayed through the deterministic engine. No AI in the scoring path.

Unlock the full CE report

Free during early access · no credit card · 1,000+ stocks tracked daily

Were we right on CE? — 70 graded days

WE MADE A CALL ON 11 OF 70 DAYS — RIGHT 2 · WRONG 9

Why so much grey? The score only takes a side when enough of the 13 factors agree on a direction. On CE they have mostly pointed different ways and cancelled each other out, so the honest output was “no call” — the model doesn't manufacture an opinion to have one. Nothing is missing here: across all 1,001 stocks about a third of graded days carry a call; this stock just hasn't been one of them lately.

Each bar is what CE actually did over the 7 days after one of our published scores — up bars, the stock rose; down bars, it fell. A “call” is a day the score left the middle zone (above 60 = we said up, below 40 = we said down), and it's counted right only if the stock then moved our way. Faint bars are days we made no call — saying “we don't know” is allowed here. These are raw counts on one stock, far too few to prove skill either way — the serious, pre-registered test across all 1,001 stocks is the public bet.