Market intelligence市场情报
Read the move. Test the cause. Follow the evidence.读懂变化，检验原因，沿证据深入。
Start an investigation开始一项研究
Fixed research lenses · no opportunity ranking固定研究视角 · 非机会排名
Read the whole market纵览商品市场
One-month price changes · select a commodity一个月价格变化 · 选择品种
Price coverage is broader than physical coverage. An index reading is not an instruction for every commodity.价格覆盖范围大于实物证据覆盖范围；指数读数不等于每个品种的操作指令。
Commodity research商品研究
The one-month move is negative; the published long-term trend remains up. A reversal is not established by this combination alone.一个月变化为负，已发布的长期趋势仍向上；仅凭这一组合，不能确认趋势反转。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
China premium · close-based proxy中国溢价 · 收盘代理指标
Source observation来源观察 · 2026-10-20 07:30 UTC · Official benchmark unavailable官方基准不可用
This regional relative price can coexist with a rising or falling global gold price. Demand, import constraints, supply and market structure can all influence the gap; its sign alone does not measure global physical demand.这一区域相对价格可以与全球金价上涨或下跌并存。需求、进口约束、供应与市场结构均会影响价差，其正负本身不能衡量全球实物需求。
Regional relative pricing does not resolve the separate rate readings or identify the cause of gold's price move.区域相对定价不能解决独立利率读数的分歧，也不能确定金价变化的原因。
Shanghai Au99.99 in CNY per gram is converted to CNY per troy ounce and compared with aligned global XAU/CNY at the Shanghai close. This is an indicative proxy, not the official Shanghai/London benchmark.上海Au99.99人民币/克价格换算为人民币/金衡盎司后，与上海收盘时钟对齐的全球XAU/CNY比较。这是参考代理指标，并非官方上海/伦敦基准。
Published history of the same comparison method.同一比较方法的已发布历史。
The source observation and the commodity price snapshot have separate clocks. A publisher freshness flag does not establish historical release availability or a simultaneously executable spread.来源观察与商品价格快照使用不同的时钟。发布来源的新鲜度标记不能证明历史发布可得性，也不表示价差可同时成交。
Commodity research商品研究
The one-month decline agrees with the published long-term trend. Supply, demand and positioning still need separate investigation.一个月跌幅与已发布的长期趋势一致，仍需分别研究供应、需求与持仓。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month decline agrees with the published long-term trend. Supply, demand and positioning still need separate investigation.一个月跌幅与已发布的长期趋势一致，仍需分别研究供应、需求与持仓。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month move is negative; the published long-term trend remains up. A reversal is not established by this combination alone.一个月变化为负，已发布的长期趋势仍向上；仅凭这一组合，不能确认趋势反转。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month advance agrees with the published long-term trend. This alignment does not identify the cause or establish a future return.一个月涨幅与已发布的长期趋势一致；这种一致性不能识别原因，也不能确定未来回报。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month move is negative; the published long-term trend remains up. A reversal is not established by this combination alone.一个月变化为负，已发布的长期趋势仍向上；仅凭这一组合，不能确认趋势反转。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month move is negative; the published long-term trend remains up. A reversal is not established by this combination alone.一个月变化为负，已发布的长期趋势仍向上；仅凭这一组合，不能确认趋势反转。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month decline agrees with the published long-term trend. Supply, demand and positioning still need separate investigation.一个月跌幅与已发布的长期趋势一致，仍需分别研究供应、需求与持仓。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month advance agrees with the published long-term trend. This alignment does not identify the cause or establish a future return.一个月涨幅与已发布的长期趋势一致；这种一致性不能识别原因，也不能确定未来回报。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
US petroleum balance美国石油供需
Week ending 2026-10-16 · 13 days before the price snapshot · release vintage unavailable.截至2026-10-16的一周 · 早于价格快照13天 · 发布版本不可核实。
All three stock categories were below their own seasonal norms in this observation week. This shows a tight US inventory buffer; it does not identify the cause or establish a global supply loss.该观察周的三类库存均低于各自季节常态，说明美国库存缓冲偏紧；尚不能识别原因或确认全球供应损失。
Week ending观察周截至 2026-10-16
Week ending观察周截至 2026-10-16
Week ending观察周截至 2026-10-16
The inflation link needs its own dated evidence.通胀传导需要独立的有日期证据。
Check realised prices and hedges for producers, product spreads for refiners, and fuel pass-through for consumers.生产商需核对实现价格与套保，炼油商需核对产品价差，消费端需核对燃料成本转嫁。
Inventory changes alone do not distinguish supply, demand, refinery operations or trade.库存变化本身无法区分供应、需求、炼厂运作或贸易因素。
Stock changes use endpoints exactly 28 days apart. Cover divides stocks by average daily refinery inputs or product supplied over the complete weekly window shown below. Product supplied is a consumption proxy, not retail sales.库存变化使用相隔整28天的端点。库存覆盖天数以库存除以下方完整周度窗口中日期对齐的日均炼厂投入或产品供应量；产品供应量是消费代理指标，不是零售销量。
| Activity / supply evidence活动 / 供应证据 | Value数值 | Observation window观察窗口 |
|---|---|---|
| Refinery inputs · average炼厂投入 · 均值 | 16.5 million bbl/day百万桶/日 | 2026-09-25 → 2026-10-16 |
| Gasoline supplied · average汽油供应量 · 均值 | 8.712 million bbl/day百万桶/日 | 2026-09-25 → 2026-10-16 |
| Distillate supplied · average馏分油供应量 · 均值 | 3.913 million bbl/day百万桶/日 | 2026-09-25 → 2026-10-16 |
| US crude production美国原油产量 | 13.27 million bbl/day百万桶/日 | 2026-10-16 |
| Production change · 4 weeks产量变化 · 四周 | -0.67 million bbl/day百万桶/日 | 2026-09-18 → 2026-10-16 |
| Refinery utilisation炼厂利用率 | 93.4 % | 2026-10-16 |
| SPR stock change · 4 weeks战略储备库存变化 · 四周 | -1.87 million bbl百万桶 | 2026-09-18 → 2026-10-16 |
| Cushing crude stocks库欣原油库存 | 27.6 million bbl百万桶 | 2026-10-16 |
Seasonal z compares each stock with its own ISO-week mean, scaled by the variation of seasonal anomalies. The existing method includes the latest observation; it is not EIA’s published five-year range. Negative means below-season stocks. The actual reference coverage is listed below.季节z值将各库存与自身ISO周均值比较，再以季节异常的波动幅度标准化。现有方法包含最新观察，并非EIA发布的五年区间；负值表示低于季节常态。实际参照覆盖如下。
| Stock series库存序列 | Seasonal reference季节参照 | Stock-change window库存变化窗口 |
|---|---|---|
| Commercial crude商业原油 | 2021-09-24 → 2026-10-16 · 265 observations个观察值 | 2026-09-18 → 2026-10-16 |
| Gasoline汽油 | 2021-09-24 → 2026-10-16 · 265 observations个观察值 | 2026-09-18 → 2026-10-16 |
| Distillate馏分油 | 2021-09-24 → 2026-10-16 · 265 observations个观察值 | 2026-09-18 → 2026-10-16 |
Release and revision timestamps are not recorded in this cache. These observations cannot confirm what was known at the earlier price snapshot. US stocks do not measure the complete global balance.该缓存未记录发布及修订时间戳；这些观察不能确认较早价格快照时已知的信息。美国库存也不代表完整的全球供需。
Commodity research商品研究
The one-month move is negative; the published long-term trend remains up. A reversal is not established by this combination alone.一个月变化为负，已发布的长期趋势仍向上；仅凭这一组合，不能确认趋势反转。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month move is negative; the published long-term trend remains up. A reversal is not established by this combination alone.一个月变化为负，已发布的长期趋势仍向上；仅凭这一组合，不能确认趋势反转。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month move is negative; the published long-term trend remains up. A reversal is not established by this combination alone.一个月变化为负，已发布的长期趋势仍向上；仅凭这一组合，不能确认趋势反转。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month move is positive; the published long-term trend remains down. A recovery needs further confirmation.一个月变化为正，已发布的长期趋势仍向下；复苏需要进一步确认。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month decline agrees with the published long-term trend. Supply, demand and positioning still need separate investigation.一个月跌幅与已发布的长期趋势一致，仍需分别研究供应、需求与持仓。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month advance agrees with the published long-term trend. This alignment does not identify the cause or establish a future return.一个月涨幅与已发布的长期趋势一致；这种一致性不能识别原因，也不能确定未来回报。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month advance agrees with the published long-term trend. This alignment does not identify the cause or establish a future return.一个月涨幅与已发布的长期趋势一致；这种一致性不能识别原因，也不能确定未来回报。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Commodity research商品研究
The one-month move is positive; the published long-term trend remains down. A recovery needs further confirmation.一个月变化为正，已发布的长期趋势仍向下；复苏需要进一步确认。
Research dated研究日期 · 2026-10-29How the move changes with the window观察窗口如何影响涨跌读数
Overlapping lookbacks · shared scale · not a price path重叠回看窗口 · 统一刻度 · 非价格轨迹
These returns come from the existing published price series. The one-month change and trading-session windows are distinct comparisons. No dated futures contract, delivery point or executable quote is inferred.收益率来自现有已发布价格序列。一个月变化与交易日窗口属于不同的比较口径。本页不据此推定具体期货合约、交割地点或可成交报价。
Explanation to test待检验的解释
Physical evidence behind this investigation本次研究的实物证据
Week ending 2026-10-16 · 13 days before the price snapshot · release vintage unavailable.截至2026-10-16的一周 · 早于价格快照13天 · 发布版本不可核实。
The inflation link needs its own dated evidence.通胀传导需要独立的有日期证据。
China premium · close-based proxy中国溢价 · 收盘代理指标
2026-10-20 07:30 UTC · +0.31%
Regional relative pricing does not resolve the separate rate readings or identify the cause of gold's price move.区域相对定价不能解决独立利率读数的分歧，也不能确定金价变化的原因。
+12 bp · 22 trading sessions22个交易日
Earlier context较早背景 · 2026-10-20 Open Transmission Intelligence打开传导情报 ↗This dollar reading is not qualified as same-date evidence for the commodity snapshot.此美元读数未被确认为商品快照的同日证据。
Earlier context较早背景 · 2026-10-21A dollar direction is unavailable.美元方向暂不可用。 Co-movement needs its own dated method before it can be displayed here.历史联动需要独立的日期与方法信息，才能在此展示。
US 10-year nominal yield. Yield changes and bond total returns are different quantities.美国10年期名义收益率。收益率变化与债券总回报属于不同的指标。
Earlier context较早背景 · 2026-10-20 Open Bonds打开债券页 ↗The ratio belongs to separate dated context.该比率属于另一日期的背景。
The ratio belongs to separate dated context.该比率属于另一日期的背景。
Relative price is descriptive. Copper/gold does not by itself establish growth or a rate forecast; gold/silver does not imply a mean-reversion trade.相对价格用于描述。铜金比不能单独确定经济增长或利率预测，金银比也不意味着均值回归交易。
Realised prices, production and hedges determine revenue exposure.实现价格、产量与套期保值决定收入敞口。
How much production is hedged?有多少产量已套保？Product prices relative to crude matter more than crude alone.成品油相对原油的价格比单独油价更重要。
Are refining margins expanding or compressing?炼油利润率扩大还是收窄？Fuel costs, hedging and customer pass-through can offset one another.燃料成本、套保与对客户的成本转嫁可能相互抵消。
How quickly can costs be passed through?成本多快能转嫁？Selling prices, volumes, mine costs and cost currencies shape margins.销售价格、产量、矿山成本及成本货币影响利润率。
Do costs rise with the gold price?成本会随金价上升吗？Contract terms and attributable production determine receipts.合约条款及权益产量决定收入。
Which operating and counterparty risks matter?哪些运营与交易对手风险重要？Prices, grades, output and by-product credits shape cash flow.价格、品位、产量及副产品抵扣影响现金流。
Is price strength offset by lower production?较高价格是否被产量下降抵消？Input costs interact with pass-through terms and end demand.投入成本与成本转嫁条款及终端需求相互作用。
How long is the pass-through lag?成本转嫁的时滞多长？A business connection needs dated company evidence for costs, volumes, hedges and contracts. It does not by itself establish stock attractiveness. The existing equity mechanism screens below are a separate starting point.企业关联需要有日期的成本、产量、套保及合约证据，不能单独决定股票吸引力。下方现有股票机制筛选提供另一研究起点。
Existing transmission sequence现有传导序列
The transmission snapshot is unavailable or newer than this research date. Its stages are withheld.传导快照不可用或晚于本次研究日期，因此不展示其环节状态。
Separate observed stages do not prove timing, causality or a tradable forecast.单独观察到的环节，不能证明先后时序、因果关系或可交易的预测。
Earlier context较早背景 · 2026-10-20 · 3 / 4 conditions met in this source; episode transitions are assessed separately个来源条件满足；历史事件传导单独评估
Separate source context. These checks and equity screens do not support the selected research date.单独的来源背景：这些条件检查及股票筛选不作为所选研究日期的支持证据。
| Source metric来源指标 | Window窗口 | Observed观察值 | Owner condition原有条件 |
|---|---|---|---|
| CL=F ret_pct | 60 | 25.745 % | gt 25 % |
| CL=F ma50_slope5 | Not supplied未提供 | 1.4764 | gt 0 |
| T10YIE ret_bp | 22 | 2.0 bp | gt 15 bp |
| state.rates.direction | Not supplied未提供 | rising | eq rising |
| state.rates.real_10y_chg_63d_bp | Not supplied未提供 | 37.0 | gt 30 |
| QQQ / SPY rs_pp | 63 | -1.304 pp | lt 0 pp |
Recorded episode已记录事件: oil_inflation_duration_derate@r2:2026-10-15. Recorded hop confirmations can belong to a past episode; they do not establish conditions today.已记录的传导确认可能属于过去的事件，不代表当前条件。
| Recorded transition已记录的传导 | Confirmed on确认日期 | Owner lag window (days)原有时滞窗口（天） |
|---|---|---|
| oil_shock->breakeven_rise | Unconfirmed or undated未确认或无日期 | [5, 60] |
| breakeven_rise->yield_rise | Unconfirmed or undated未确认或无日期 | [0, 90] |
| yield_rise->duration_derate | Unconfirmed or undated未确认或无日期 | [0, 120] |
The publisher supplies a research date, not archival proof of when every observation became available. Historical transition probabilities are not multiplied into a path forecast.发布内容提供研究日期，但不构成每条观察首次可用时间的档案证明；历史环节概率不会被相乘为路径预测。
Existing transmission sequence现有传导序列
The transmission snapshot is unavailable or newer than this research date. Its stages are withheld.传导快照不可用或晚于本次研究日期，因此不展示其环节状态。
Separate observed stages do not prove timing, causality or a tradable forecast.单独观察到的环节，不能证明先后时序、因果关系或可交易的预测。
Earlier context较早背景 · 2026-10-20 · 0 / 4 conditions met in this source; episode transitions are assessed separately个来源条件满足；历史事件传导单独评估
Separate source context. These checks and equity screens do not support the selected research date.单独的来源背景：这些条件检查及股票筛选不作为所选研究日期的支持证据。
| Source metric来源指标 | Window窗口 | Observed观察值 | Owner condition原有条件 |
|---|---|---|---|
| CL=F ret_pct | 20 | 7.294 % | lt -12 % |
| CL=F ret_pct | 60 | 25.745 % | lt -18 % |
| CL=F ma50_slope5 | Not supplied未提供 | 1.4764 | lt 0 |
| T10YIE ret_bp | 22 | 2.0 bp | lt -12 bp |
| DFII10 ret_bp | 22 | 12.0 bp | lt -12 bp |
| QQQ / SPY rs_pp | 63 | -1.304 pp | gt 0 pp |
Recorded episode已记录事件: None supplied未提供. Recorded hop confirmations can belong to a past episode; they do not establish conditions today.已记录的传导确认可能属于过去的事件，不代表当前条件。
| Recorded transition已记录的传导 | Confirmed on确认日期 | Owner lag window (days)原有时滞窗口（天） |
|---|---|---|
| oil_slide->breakeven_fall | Unconfirmed or undated未确认或无日期 | [5, 60] |
| breakeven_fall->real_yield_fall | Unconfirmed or undated未确认或无日期 | [0, 90] |
| real_yield_fall->duration_rerate | Unconfirmed or undated未确认或无日期 | [0, 120] |
The publisher supplies a research date, not archival proof of when every observation became available. Historical transition probabilities are not multiplied into a path forecast.发布内容提供研究日期，但不构成每条观察首次可用时间的档案证明；历史环节概率不会被相乘为路径预测。
Existing transmission sequence现有传导序列
The transmission snapshot is unavailable or newer than this research date. Its stages are withheld.传导快照不可用或晚于本次研究日期，因此不展示其环节状态。
Separate observed stages do not prove timing, causality or a tradable forecast.单独观察到的环节，不能证明先后时序、因果关系或可交易的预测。
Earlier context较早背景 · 2026-10-20 · 1 / 3 conditions met in this source; episode transitions are assessed separately个来源条件满足；历史事件传导单独评估
Separate source context. These checks and equity screens do not support the selected research date.单独的来源背景：这些条件检查及股票筛选不作为所选研究日期的支持证据。
| Source metric来源指标 | Window窗口 | Observed观察值 | Owner condition原有条件 |
|---|---|---|---|
| state.rates.real_10y_pctile | Not supplied未提供 | 0.99 | gte 0.9 |
| state.rates.regime | Not supplied未提供 | restrictive | eq restrictive |
| DFII10 ret_bp | 22 | 12.0 bp | lt -12 bp |
| GC=F ret_pct | 22 | -4.124 % | gt 4 % |
Recorded episode已记录事件: real_rate_peak_gold_rerate@r1:2026-10-14. Recorded hop confirmations can belong to a past episode; they do not establish conditions today.已记录的传导确认可能属于过去的事件，不代表当前条件。
| Recorded transition已记录的传导 | Confirmed on确认日期 | Owner lag window (days)原有时滞窗口（天） |
|---|---|---|
| real_rate_extreme->real_rate_rolldown | Unconfirmed or undated未确认或无日期 | [0, 180] |
| real_rate_rolldown->gold_bid | Unconfirmed or undated未确认或无日期 | [0, 90] |
The publisher supplies a research date, not archival proof of when every observation became available. Historical transition probabilities are not multiplied into a path forecast.发布内容提供研究日期，但不构成每条观察首次可用时间的档案证明；历史环节概率不会被相乘为路径预测。
These are existing mechanism screens. Membership is not a verified direct commodity exposure or an investment ranking.这些是现有机制筛选结果；名单归属不代表已验证的商品直接敞口或投资排名。
long-duration names whose discount-rate headwind releases as real yields roll down贴现率逆风随实际收益率回落而释放的长久期名字
689 names in the source screen个来源筛选名单 · 276 not assessable个无法评估
6 / 687 identity-verified names shown alphabetically, not ranked个身份已验证名称，按字母顺序展示，非排名
A dated commodity-specific transmission sequence is not qualified for this selection. Inspect the available price evidence, physical coverage and counterpart markets before applying an explanation.本品种尚无合格且有日期的专属传导序列。采用一种解释前，请核对价格证据、实物市场覆盖及关联市场。
A shared perspective统一的观察视角
Choose up to four commodities最多选择四个品种
Select up to four commodities above to compare the same observation window.请在上方选择最多四个品种，比较同一观察窗口。
One zero baseline. One scale. Missing observations remain unavailable.共用零基线与刻度；缺失观察仍显示为不可用。
| Commodity品种 | Price change价格变化 | Long-term trend长期趋势 | Coverage覆盖范围 |
|---|---|---|---|
| -4.3% | Up向上 | Price + explanation tests价格与解释检验 | |
| -14.4% | Down向下 | Price observations价格观察 | |
| -13.0% | Down向下 | Price observations价格观察 | |
| -14.5% | Up向上 | Price observations价格观察 | |
| +1.4% | Up向上 | Price + explanation tests价格与解释检验 | |
| -8.7% | Up向上 | Price observations价格观察 | |
| -1.3% | Up向上 | Price observations价格观察 | |
| -2.4% | Down向下 | Price observations价格观察 | |
| +10.5% | Up向上 | Price + explanation tests价格与解释检验 | |
| -10.2% | Up向上 | Price observations价格观察 | |
| -3.9% | Up向上 | Price observations价格观察 | |
| -11.6% | Up向上 | Price observations价格观察 | |
| +0.9% | Down向下 | Price observations价格观察 | |
| -1.0% | Down向下 | Price observations价格观察 | |
| +4.5% | Up向上 | Price observations价格观察 | |
| +9.5% | Up向上 | Price observations价格观察 | |
| +4.2% | Down向下 | Price observations价格观察 |
These are price-series comparisons, not total-return rankings. Futures rolls, currencies and contract selection can change an investor’s result.这是价格序列比较，并非总回报排名。期货展期、汇率与合约选择可能改变投资者的实际结果。
Your research worksheet您的研究工作页
Portable draft · monitoring is not active可导出的草稿 · 未启用监测
| Publisher发布来源 | Research date研究日期 | Use in this snapshot在本快照中的使用 |
|---|---|---|
| data/commodity/complex_latest.json | 2026-10-29 | Research dated研究日期 2026-10-29 |
| data/transmission/latest.json | 2026-10-20 | Earlier context较早背景 · 2026-10-20 |
| data/forex/latest.json | 2026-10-21 | Earlier context较早背景 · 2026-10-21 |
| data/transmission/chain_state.json | 2026-10-20 | Earlier context较早背景 · 2026-10-20 |
| data/commodity/latest.json | 2026-10-21 | Earlier context较早背景 · 2026-10-21 |
| data/eia/ | 2026-10-16 | Earlier context较早背景 · 2026-10-16 |
| data/commodity/latest.json | 2026-10-20 | Earlier context较早背景 · 2026-10-20 |
Research snapshots are not live quotes. Publisher dates do not establish point-in-time publication availability. Conditional explanations are not causal identification or investment signals.研究快照不是实时报价。发布方研究日期不构成信息首次可用时间的证明；条件性解释也不等于因果识别或投资信号。
Oil explanations distinguish supply and demand mechanisms; weekly product-demand evidence should be read with seasonal and multi-week context.原油解释区分供给与需求机制；每周产品需求证据应结合季节性及多周背景理解。 Kilian · oil shock researchKilian · 原油冲击研究 · EIA
mmx.commodities.workspace.v1 · fe4e36699b9d21d3516d3d63db1235d25cd9f1d675a5df06c65fa7ea9addacbc
These are the existing publishers’ detailed readings. Their dates, horizons and model judgements remain distinct from the research interpretation above.以下为现有发布来源的详细判断；其日期、时间窗口及模型结论，与上方研究解释分别呈现。
Where commodities stand today大宗商品今日处境
How many agree多少品种共振
Commodity index — the whole complex大宗商品指数——整个板块
The long cycle — where each commodity sits长周期——各品种所处位置
signals lining up — oversold/overbought, shock, positioning, cycle引擎信号汇总——超卖/超买、冲击、持仓、周期
watch-list, not a trade trigger观察清单，非交易触发1-month move · short-term momentum · long cycle1月涨跌 · 短期动量 · 长周期
select a commodity to see its full read选择品种查看完整分析
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | up↑ | 60.0 | 34.0 | turning down转折向下 |
| 3-Day3日 | up↑ | 61.0 | 81.0 | turning down转折向下 |
| Weekly周线 | up↑ | 60.0 | 97.0 | +正 |
| Biweekly双周 | up↑ | 59.0 | 50.0 | +正 |
| Monthly月线 | up↑ | 62.0 | 61.0 | +正 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 48.0 | 42.0 | −负 |
| 3-Day3日 | down↓ | 48.0 | 20.0 | turning down转折向下 |
| Weekly周线 | up↑ | 48.0 | 47.0 | +正 |
| Biweekly双周 | up↑ | 47.0 | 78.0 | turning up转折向上 |
| Monthly月线 | down↓ | 47.0 | 11.0 | −负 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 44.0 | 53.0 | −负 |
| 3-Day3日 | down↓ | 46.0 | 12.0 | −负 |
| Weekly周线 | down↓ | 49.0 | 6.0 | −负 |
| Biweekly双周 | down↓ | 54.0 | 5.0 | turning down转折向下 |
| Monthly月线 | up↑ | 57.0 | 67.0 | +正 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 43.0 | 17.0 | turning down转折向下 |
| 3-Day3日 | down↓ | 52.0 | 21.0 | −负 |
| Weekly周线 | up↑ | 56.0 | 36.0 | turning down转折向下 |
| Biweekly双周 | up↑ | 59.0 | 49.0 | turning down转折向下 |
| Monthly月线 | up↑ | 66.0 | 86.0 | turning down转折向下 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 35.0 | 54.0 | −负 |
| 3-Day3日 | down↓ | 37.0 | 9.0 | −负 |
| Weekly周线 | down↓ | 39.0 | 10.0 | −负 |
| Biweekly双周 | down↓ | 45.0 | 4.0 | −负 |
| Monthly月线 | flat→ | 55.0 | 6.0 | −负 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 27.0 | 12.0 | −负 |
| 3-Day3日 | down↓ | 31.0 | 0.0 | −负 |
| Weekly周线 | down↓ | 36.0 | 4.0 | −负 |
| Biweekly双周 | down↓ | 42.0 | 1.0 | −负 |
| Monthly月线 | down↓ | 50.0 | 5.0 | turning down转折向下 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 27.0 | 8.0 | −负 |
| 3-Day3日 | down↓ | 33.0 | 0.0 | −负 |
| Weekly周线 | down↓ | 33.0 | 3.0 | turning down转折向下 |
| Biweekly双周 | down↓ | 41.0 | 7.0 | −负 |
| Monthly月线 | down↓ | 49.0 | 9.0 | −负 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 22.0 | 0.0 | −负 |
| 3-Day3日 | down↓ | 26.0 | 0.0 | −负 |
| Weekly周线 | down↓ | 30.0 | 0.0 | −负 |
| Biweekly双周 | down↓ | 36.0 | 0.0 | −负 |
| Monthly月线 | down↓ | 41.0 | 8.0 | turning down转折向下 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | up↑ | 51.0 | 85.0 | turning up转折向上 |
| 3-Day3日 | down↓ | 52.0 | 22.0 | −负 |
| Weekly周线 | up↑ | 58.0 | 8.0 | turning up转折向上 |
| Biweekly双周 | down↓ | 65.0 | 60.0 | turning down转折向下 |
| Monthly月线 | up↑ | 69.0 | 97.0 | +正 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 30.0 | 30.0 | −负 |
| 3-Day3日 | down↓ | 40.0 | 0.0 | −负 |
| Weekly周线 | down↓ | 45.0 | 2.0 | turning down转折向下 |
| Biweekly双周 | up↑ | 49.0 | 48.0 | +正 |
| Monthly月线 | flat→ | 49.0 | 75.0 | turning down转折向下 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 24.0 | 5.0 | turning up转折向上 |
| 3-Day3日 | down↓ | 37.0 | 1.0 | −负 |
| Weekly周线 | down↓ | 42.0 | 0.0 | −负 |
| Biweekly双周 | down↓ | 48.0 | 29.0 | turning down转折向下 |
| Monthly月线 | up↑ | 51.0 | 66.0 | +正 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 36.0 | 12.0 | turning up转折向上 |
| 3-Day3日 | down↓ | 48.0 | 6.0 | −负 |
| Weekly周线 | down↓ | 52.0 | 25.0 | turning down转折向下 |
| Biweekly双周 | up↑ | 56.0 | 59.0 | +正 |
| Monthly月线 | up↑ | 56.0 | 92.0 | turning down转折向下 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 53.0 | 29.0 | turning down转折向下 |
| 3-Day3日 | up↑ | 49.0 | 87.0 | +正 |
| Weekly周线 | up↑ | 47.0 | 94.0 | turning up转折向上 |
| Biweekly双周 | down↓ | 48.0 | 27.0 | −负 |
| Monthly月线 | down↓ | 52.0 | 4.0 | −负 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 40.0 | 5.0 | turning down转折向下 |
| 3-Day3日 | down↓ | 43.0 | 48.0 | −负 |
| Weekly周线 | down↓ | 44.0 | 31.0 | −负 |
| Biweekly双周 | down↓ | 45.0 | 48.0 | turning down转折向下 |
| Monthly月线 | down↓ | 47.0 | 25.0 | −负 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | flat→ | 49.0 | 11.0 | turning up转折向上 |
| 3-Day3日 | down↓ | 61.0 | 16.0 | turning down转折向下 |
| Weekly周线 | up↑ | 66.0 | 77.0 | +正 |
| Biweekly双周 | up↑ | 62.0 | 92.0 | turning down转折向下 |
| Monthly月线 | up↑ | 59.0 | 99.0 | turning up转折向上 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 43.0 | 10.0 | −负 |
| 3-Day3日 | down↓ | 49.0 | 4.0 | −负 |
| Weekly周线 | down↓ | 52.0 | 11.0 | turning down转折向下 |
| Biweekly双周 | up↑ | 50.0 | 82.0 | +正 |
| Monthly月线 | down↓ | 48.0 | 79.0 | −负 |
What to do now当前该怎么做
Do the signals agree?各周期是否共振？
| Timeframe周期 | Trend趋势 | Momentum动量 | Stretch拉伸 | Turn转折 |
|---|---|---|---|---|
| Daily日线 | down↓ | 45.0 | 11.0 | turning down转折向下 |
| 3-Day3日 | flat→ | 48.0 | 38.0 | turning up转折向上 |
| Weekly周线 | down↓ | 51.0 | 23.0 | −负 |
| Biweekly双周 | down↓ | 55.0 | 13.0 | turning down转折向下 |
| Monthly月线 | up↑ | 53.0 | 66.0 | +正 |
Five commodity families. Where we read physical supply, and where we do not yet.五大商品族：哪些能读到实物供需，哪些还没有。
engine/commodity_inputs.py · Daily price history (Yahoo)每日价格历史（Yahoo）engine/commodity_supply_context.py · EIA Weekly Petroleum Status ReportEIA 每周石油状况报告engine/commodity_inputs.py · Daily price history (Yahoo)每日价格历史（Yahoo）engine/commodity_inputs.py · Daily price history (Yahoo)每日价格历史（Yahoo）engine/commodity_inputs.py · Daily price history (Yahoo)每日价格历史（Yahoo）engine/critical_minerals_supply.py · USGS Mineral Commodity Summaries美国地质调查局《矿产商品摘要》Coverage is read from this build, not hand-listed. A balance read, never a price call.覆盖状态由本次构建实时读取，非人工登记。仅为供需背景，不构成价格判断。
Three industry stages, measured in the same week. These are aggregate quantities, not traced movements of the same barrels.同一周的三个产业环节。这些是行业总量，并非同一批原油的流向追踪。
US crude field production美国原油田产量: 13,269
Refinery crude inputs炼油厂原油投入量: 16,649
Includes imported crude; it is not a subset of US field production.包含进口原油，并非美国原油产量的子集。
Gasoline supplied汽油供应量: 8,982 · Distillate supplied馏分油供应量: 3,732
Products supplied approximates consumption through disappearance from primary supplies. These two products are not the entire output.产品供应量以初级供应环节的流出量近似消费量。这两种产品并非全部产出。
Annual physical-supply context, not a live shortage indicator or a position recommendation.年度实物供应背景，不代表实时短缺，也不构成仓位建议。
Congo (Kinshasa) mined about 74% of the world's cobalt in 2025, and the US imported about 79 percent of what it used.刚果（金）在2025年开采了全球约74%的钴，美国进口了其用量的约79%。
Observation period观测期间: 2025
Australia mined about 32% of the world's lithium in 2025, and the US imported more than 50 percent of what it used.澳大利亚在2025年开采了全球约32%的锂，美国进口了其用量的超过50%。
Observation period观测期间: 2025
China produced about 100% of the world's gallium in 2025, and the US imported about 100 percent of what it used.中国在2025年生产了全球约100%的镓，美国进口了其用量的约100%。
Observation period观测期间: 2025
China mined about 69% of the world's rare earths in 2025, and the US imported about 67 percent of what it used.中国在2025年开采了全球约69%的稀土，美国进口了其用量的约67%。
Observation period观测期间: 2025
USGS Mineral Commodity Summaries data release美国地质调查局《矿产商品摘要》数据发布