牧原股份

002714 Research Preview

Confirmed: The oversold cluster is consistent. RSI14 (26.55), RSI6 (24.06), KDJ in the high-20s, and price pinned near the Bollinger lower band (32.61) all point to a stretched downside condition with rebound potential, which is why the RSI extreme sub-score reads 75/100 (Buy) and Bollinger reads 60/100 (Buy).

Conflicted: Momentum signals disagree with flow. The MACD histogram has turned slightly positive (+0.11) and the precomputed context calls this a "Bullish MACD configuration," yet both MACD lines remain well below zero (-1.86 / -1.97) and the MACD sub-score is only 50/100, with the note that momentum is weakening. Against this, OBV (slope -152.19%) and the volume-price sub-score (35/100, Sell, price-down-on-volume ~1.4x) show active selling. So a possible momentum turn is fighting visible distribution.

Also note an internal inconsistency: the supplied moving-average label reads "Mixed alignment (MA5 > MA10 > MA20 > MA60)," but the raw values show MA10 (34.55) below MA20 (36.01), and the MA-cross sub-score (40/100) explicitly flags MA5<MA20 死叉. We treat the alignment as bearish-to-mixed, not cleanly stacked, and flag the label as lower confidence.

Missing: No custom indicators were supplied beyond the standard set, so none are folded in here.

Note: PEG proxy, the fair-value range, the base fair-value gap, and the 5Y growth forecast are StockKit scenario-model outputs computed from public financial history (income period 2025-12-31). They are internal forecasts, not analyst consensus targets. The model's Base fair value (47.49) sits above the current 33.63, implying a +41.2% gap, but the negative net margin (-4.09%) and thin gross margin (+5.17%) are a tension to weigh against that gap. EV/EBITDA, dividend yield, market cap, and 52-week range were not supplied and are omitted.

Because no headlines are available, this brief carries no event-driven view. There is nothing in the supplied data to explain the -2.97% session or the visible OBV distribution, so any catalyst attribution would be unsupported. Treat the absence as a gap, not as a quiet tape.

No provider sentiment series was returned. Social channels are not connected, so no Reddit, X, StockTwits, or local-forum readings are inferred. Sentiment is effectively a blank panel here and should not be weighted in the thesis.

1. 32.61 support / 33.52 30-session low: confirm whether the shelf holds or breaks (highest priority). 2. Daily volume vs 20-session average (currently +38.74%) and OBV slope (-152.19%) for signs distribution is easing. 3. MA5 (34.70) and MA10 (34.55) as first reclaim levels for any rebound. 4. RSI14 (26.55) / RSI6 (24.06): look for an upturn out of oversold rather than further depression. 5. MACD histogram (+0.11): does the nascent positive turn hold or fade back negative. 6. MA20 (36.01) as the line separating consolidation from a real recovery; resistance 39.42 above. 7. News and sentiment feeds: re-poll for any new headline, since both panels were empty this cycle (lower-confidence area).

This brief is a public-facing research preview built only from the supplied StockKit dataset. It is not personalized investment advice and makes no return promises. Confidence is medium-low overall: technical and valuation inputs are populated, but the news and sentiment feeds returned no data, and the moving-average alignment label conflicts with the raw MA values, so those areas are flagged as lower confidence. Levels cited are watch levels for monitoring, not instructions to act. StockKit scenario-model rows are internally computed forecasts, not analyst consensus.