ψ Psi Quant Research

Signal Report · US

TXN HOLD

Composite decision report grounded in 1-XH strategy signals and a completed multi-agent debate (agents-3node-v1), generated at 2026-08-14 09:29.

Final Agent Decision

Action
HOLD
Weight
0%
Confidence
55%
Variant
xiaohan_gogo_v2

Reasons

Bull Case (Analyst)

- Technical setup is constructive: the 1-XH family prints a buy with score 0.60, and the gogo_detail shows bottoms forming across 1h–4h timeframes. Price at $279.6 sits above the fast EMA ($277) and is pressing toward the slow EMA ($281.7), suggesting momentum is building from a multi-timeframe base.
- Cross-sectional factor view corroborates the technical signal: TXN ranks 50/252 (top 19.8%) with composite z = +0.35 in a BULL regime catalog (34 factors, zero direction-flips). The strongest single contributor is xh_gogo_score at +3.00 — the same family as the technical signal — meaning the factor model independently confirms the momentum setup.
- Q2 earnings delivered a beat: EPS of $2.14 vs. $1.92 consensus, revenue of $5.46B, and management cited broad demand across industrial, automotive, and data centre customers. The market's initial reaction (stock +3.39% on Jul 21) suggests the inflection point narrative is gaining traction.
- The $7.8B acquisition of Silicon Laboratories (announced Feb 2026) expands TI's analog portfolio; combined with the earnings beat, this gives a fundamental catalyst beyond pure technicals.
- Beta_60d (+0.76 z) is a positive contributor — in a BULL regime, higher-beta names tend to outperform as risk appetite expands.

Bear Case (Analyst)

- The cross-sectional rank is only 50/252 (top 19.8%), and TXN is NOT in the production basket — the model explicitly did not place it in the top decile this period. The composite z of +0.35 is modest, not a high-conviction long.
- The bear case is driven by three negative factors: close_to_high_60d (-1.40 z) shows the stock is far from its 60-day high despite the recent rally; sharpe_60d (-0.79 z) indicates poor risk-adjusted returns over the past two months; mean_revert_60d (-0.64 z) suggests the stock is stretched and due for a pullback.
- Despite the earnings beat, shares "slip" and "slip even as Q2 results, guidance top estimates" per two separate headlines — the market sold the news, and Goldman Sachs maintains a "sell" rating with a $225 price target (20% downside from $279.6).
- The technical signal is weak-to-moderate: score 0.60 is barely above the buy threshold, and price ($279.6) is still below the slow EMA ($281.7). The "bottom" formation across 1h–4h could easily fail if the stock cannot reclaim the slow EMA.
- The acquisition of Silicon Labs for $7.8B in cash adds integration risk and balance-sheet strain at a time when the stock is already underperforming on a risk-adjusted basis (sharpe_60d = -0.79).

Devil's Advocate

COUNTER TO BULL:
- The factor model "confirms" the technical signal only because both use the same xh_gogo family — that's circular, not independent. You're double-counting one input.
- A composite z of +0.35 is statistically indistinguishable from zero. Ranking 50/252 in a BULL regime with 34 factors means the model is lukewarm, not constructive.
- The earnings beat is stale — the stock already reacted (+3.39%) on Jul 21. You're anchoring on a past catalyst while price now sits below the slow EMA, meaning the market has already digested it.
- Beta_60d at +0.76 z is a risk factor, not a return factor. In a BULL regime it helps, but if the regime wobbles, this same beta becomes the largest drag — you're implicitly forecasting regime persistence with no evidence.
- The Silicon Labs deal is a $7.8B cash outlay. Acquisitions in analog are notoriously hard to integrate; you're treating a potential value-destroying use of cash as a catalyst without any margin or EPS accretion analysis.

COUNTER TO BEAR:
- The "sell" from Goldman with a $225 target is a single sell-side opinion, not a market signal. If the stock is at $279.6, the market is explicitly rejecting that target — citing it as bearish evidence is cherry-picking.
- The three negative factors (close_to_high, sharpe, mean_revert) are all 60-day lookbacks. They describe the past, not the future. If the stock is forming a base (as the technicals suggest), these factors will mechanically flip positive within weeks — they are lagging, not leading.
- "Market sold the news" on earnings is a weak claim without volume/flow context. A +3.39% move on Jul 21 followed by a modest drift is consistent with profit-taking, not distribution.
- Score 0.60 is above the buy threshold — that's the definition of a signal. Calling it "barely above" is arbitrary; the threshold exists precisely to separate signal from noise, and it cleared it.
- The bear case ignores the cross-sectional regime entirely. In a BULL regime catalog, a top-quintile rank with zero direction-flips is not a short — it's a "don't add" at worst. The bear thesis needs a regime shift to work, and nothing in the data suggests one.

Technical Signal (1-XH)

action
buy
score
0.60
variant
xiaohan_gogo_v2
timestamp
2026-08-14T15:30:00-04:00
close
279.6
ema_low_fast
277
ema_high_slow
281.7
gogo_score
0.6
gogo_detail
1h_bottom,2h_bottom,3h_bottom,4h_bottom

📈 Backtest Track Record xiaohan_gogo_v2 · signals simulated over available history (~3-month 30-min window, accumulating daily)

Trades
3
Win rate
67%
Avg / trade
+18.4%
Total (compounded)
+55%
best / worst
+56.2% / -1.9%
avg holding
28.0 days
open position
yes — marked-to-market +0.9%
Entry Exit Entry Exit Return Exit reason
2026-08-12 open 277.09 279.64 +0.9% open_position_mark_to_market
2026-08-04 2026-08-12 283.40 278.04 -1.9% adaptive_breakdown_1h
2026-03-20 2026-06-02 192.52 300.70 +56.2% adaptive_breakdown_1h

复盘 = each past xiaohan_gogo_v2 buy signal simulated: enter next bar, manage adds / adaptive sells / −25% hard stop. Demonstration only, not investment advice.

News & social (7 items)