# 每日结算与期货保证金 用同终点的两条价格路径重建VM、担保余额、外部现金和追缴缺口。 Entry: zh-m11 | Node: M11 | Language: zh | Editorial revision: 2026-09-21 ## Teaching instructions 你正在带读《每日结算与期货保证金》（M11，2026-09-21-MD-review-v3）。对象为有微积分/线性代数/基础概率的高年级本科至研究生。 本次任务：逐阶段算实际转入与未满足金额，证明资金不足时不能继续实现后续收益。 先确认选择核心或研究分支。必须实际读取required_readings的指定完整单元，核标题、版本、页/节、必要表图；图像未读不得猜。运行日志起始为空，记录本次实际范围和支持的一步，不把编辑端reading_log当成自己已读。选研究分支时对应optional单元转为必读。只能取得目录/摘要/错误页时继续找正式可读入口；仍缺关键原文就说明具体缺口，不凭印象补课。 先让学习者完成一项完整诊断，而非逐个考四则运算。已会内容跳过；在同一原件与同源输入上推进： 先给路径A一张完整空白资金表，让学习者自己填VM、转入前、要求、实际、尚欠、余额与外部现金。专门追问6,500等号和73.49时为何应补1,510。再切IM/MM变式：阶段2起变成10,000/8,500，阶段3部分转入后8,800仍欠1,200；此处必须停止，不能显示反弹后终点。最后让其解释现金足够但赶不上D_i与现金不足有何不同；检验M+C=初始总现金+已处理VM。BIS支线被选择时，先读完整BoxA，分清五日口径与这个客户教学账。 核每个金额的主体、单位、时点和身份；事实、由事实推算与教学条件分别说清。读者完成后须能独立解释与迁移，不能只报正确数字。逐题使用正文完整解析反馈，必要时把失败条件放回算法；未建立融资/履约路径不输出已经实现的收益。公共包不假定能访问用户硬盘。 Before substantive teaching, actually retrieve every required reading unit for the selected scope. Read its complete designated section, including necessary assumptions, tables and footnotes. A working URL or an editorial access date is not a runtime reading receipt. Record the actual version, location, scope and what it supports. If unavailable, use a previously verified equivalent source; if the required unit remains unavailable, identify that gap rather than teach it from memory. Start runtime_reading_log empty. Once reading is complete, use a substantive diagnostic or follow the reader's request for direct explanation. Advance one complete reasoning task at a time; skip mastered basics. Distinguish original facts, supplied teaching assumptions and inference. ## Required readings and runtime protocol ```json { "export_mode": "public", "content_version": "2026-09-21-MD-review-v3", "selected_branch": "core", "learning_task": "逐阶段算实际转入与未满足金额，证明资金不足时不能继续实现后续收益。", "required_readings": [ { "source_id": "MD-S05", "title": "Money Calculations for CME-cleared Futures and Options", "authors": [ "CME Group" ], "version": "updated 2015-06-11", "access": { "kind": "pdf_full_text", "uri": "https://www.cmegroup.com/clearing/files/CME-Money-Calculations-Futures-and-Options.pdf" }, "required_unit": { "locator": "PDF pp1–3 Normal Futures Variation", "scope": "日初头寸和当日成交、现金符号与normal variation算法完整单元；无关分数报价错误不采用", "purpose": "先按正确时段算VM，再进入资金账户" }, "supports": "当日成交与期初头寸的结算现金及正常舍入；不采用无关分数报价例的印刷错误，不提供CL保证金水平。" }, { "source_id": "MD-S05-MARGIN", "title": "The Benefits of Futures Margins", "authors": [ "CME Group" ], "version": "online accessed 2026-09-21", "access": { "kind": "html_full_text", "uri": "https://www.cmegroup.com/education/courses/understanding-the-benefits-of-futures/the-benefits-of-futures-margins" }, "required_unit": { "locator": "两组保证金正文", "scope": "完整读initial/maintenance及客户资金安排相关内容", "purpose": "机制与本篇8,000/6,500教学水平分开" }, "supports": "保证金是应维持的担保余额而非货款；补足规则因制度/经纪商而异。不给百分比当作当前CL参数。" }, { "source_id": "MD-S03", "title": "WTI Interim Staff Report: Trading leading up to, on and around April 20, 2020", "authors": [ "CFTC staff" ], "version": "2020-11-23 interim report", "access": { "kind": "pdf_full_text", "uri": "https://www.cftc.gov/media/5296/InterimStaffReportNYMEX_WTICrudeOil/download" }, "required_unit": { "locator": "III.B–C；printed pp6–10及p11价格输入", "scope": "市场/合约/结算的完整背景单元及采用价格定位", "purpose": "真实历史VM与合成IM/MM分开" }, "supports": "历史May结算与May−June价差；June由两者相减得到。报告不识别负价唯一根因；未采用April2020保证金。" } ], "optional_readings": [ { "source_id": "MD-S08", "title": "Market turbulence and soaring margins: lessons from two recent episodes — Box A", "authors": [ "Benjamin H Cohen", "Kevin Tracol" ], "version": "BIS Quarterly Review, March 2023", "access": { "kind": "pdf_full_text", "uri": "https://www.bis.org/publications/perceptions-risk-and-policy-outlook-drive-markets_1.pdf" }, "required_unit": { "locator": "Box A printed pp5–6，Graph A1/脚注", "scope": "两页全文，设置、两类假想IRS、模型比较、五日VM/IM口径和结论", "purpose": "用模型比较说明VM/IM与及时现金需求；不复刻数据、不冒充公司CSA" }, "supports": "假想一年期USD付固定与GBP收固定IRS；VM五日累计/IM五日closeout，讨论模型重校准与流动性。未复现Clarus曲线，不替代AECOM CSA或CL客户账户。", "branch": "margin-liquidity", "required_if_selected": true } ], "runtime_reading_log": [], "supplied_inputs": { "content_version": "2026-09-21-MD-review-v3", "experiment_ids": [ "EXP-MD-M11-MARGIN" ], "data": { "content_version": "2026-09-21-MD-review-v3", "cutoff": "2026-09-21", "frozen": { "schema": "m-d-primary-inputs-v1", "verified_on": "2026-09-21", "money_unit": "USD", "price_unit": "USD per U.S. barrel", "contract": { "code": "CL", "quantity_barrels": "1000", "tick_usd_per_barrel": "0.01", "source_id": "MD-S02", "locator": "200102.B–C", "version": "online rule retrieved 2026-09-21; amendment effective date not supplied" }, "historical_window": { "identity": "real_observations_with_derived_june_settlement", "spread_definition": "May settlement minus June settlement", "basis_definition": "EIA Cushing daily close minus May settlement; not synchronous executable quotes", "rows": [ { "date": "2020-04-17", "may_settlement": "18.27", "may_minus_june": "-6.76", "spot": "18.31" }, { "date": "2020-04-20", "may_settlement": "-37.63", "may_minus_june": "-58.06", "spot": "-36.98" }, { "date": "2020-04-21", "may_settlement": "10.01", "may_minus_june": "-1.56", "spot": "8.91" } ], "vm_position": "teaching long one May entered at Apr17 settlement; not an actual account", "missing": [ "April2020 IM/MM", "broker and intraday calls", "synchronous executable quotes" ] }, "margin_paths": { "identity": "teaching_assumptions_not_historical_CME_margins", "contracts": "1", "direction": "long", "starting_total_cash": "12000", "initial_margin": "8000", "maintenance_margin": "6500", "call_rule": "after variation, strict balance ## 一、同一份价格风险，为什么要每天付钱 假设一张多头的亏损一直留到到期才处理。若中间累积了很大亏损，另一方必须相信它到时有能力支付。逐日结算把已经发生的价格变化分段结清，减少未结金额长期堆积；与此同时，亏损方需要在制度要求的时点准备现金。后来的价格反弹，不能拿来支付今天已经到期的要求。 CME的正常期货变动现金算法区分当日成交和日初已有头寸：前者从成交价算到当日结算价，后者从前一结算价算到当日结算价。[^MD-S05] 所以“昨日损益”“今天平仓的现金”“整个持有期损益”不是天然相等的三个数。 设一张CL多头一直持有，本例合约单位为$Q=1{,}000$桶，则第$i$次结算变化为$\mathrm{VM}_i=Q(F_i-F_{i-1})$。只要同一张头寸没有中断，未计费用的累计价格损益满足 $$ \sum_{i=1}^k\mathrm{VM}_i=Q(F_k-F_0). $$ 这是价格差相消的恒等式，不是资金充足的保证。左边每一笔都有日期，右边却只保留两个端点。接下来我们要把被右边省略的时间放回来。 ## 二、担保余额、损益和外部可用现金是三件事 初始保证金是建立头寸所需的账户担保余额；维持保证金给出继续持有时检查余额的界线。它们不是油价的首付款，也不是每天自动扣一次的费用。实际要求及补足方式须按适用规则和经纪商安排读取。[^MD-S05-MARGIN] 本篇采用一个**完全标明为教学设定的客户资金账**：一张CL多头，最初总现金12,000美元，其中8,000转作保证金，另有4,000随时可用。IM=8,000，MM=6,500；结算变化计入后，只有余额**严格小于**MM才产生补回IM的要求。没有利息、手续费、提款和盘中追加。 每次收到要求的模型阶段记作$D_i$，资金必须在进入下一价格阶段以前到达。它不是某家券商的真实几点截止。界面可检验“资金虽然存在，但晚于截止”的变式；不能把晚到的钱倒填成准时履约。 为避免边算边换口径，令$M_{i-1}$为原担保余额，$C_{i-1}$为保证金账户之外的可用现金。先算 $$ \widetilde M_i=M_{i-1}+\mathrm{VM}_i, \qquad A_i=\begin{cases} \mathrm{IM}_i-\widetilde M_i,&\widetilde M_i<\mathrm{MM}_i,\\ 0,&\widetilde M_i\geq\mathrm{MM}_i. \end{cases} $$ $A_i$是要求转入的金额。若资金准时，实际转入$T_i=\min(C_{i-1},A_i)$；否则$T_i=0$。更新$M_i=\widetilde M_i+T_i$、$C_i=C_{i-1}-T_i$，剩余要求为$A_i-T_i$。尚有缺口时，我们停在这个阶段，不编造融资、平仓价格或后面的反弹收益。 转入$T_i$只是把自己的钱从一栏挪到另一栏：$M_i+C_i$不因此增减；真正改变总额的是VM和未在本例纳入的费用等项目。把追加资金又当一次亏损，账就会重复。 ## 三、走完两条同终点路径 路径A的教学结算价为75→73.5→74.2→71.8→75.6。每个价格都是合成输入，不是历史CL行情。 | 阶段 | 结算价，美元/桶 | VM，美元 | 转入前余额 | 要求/实际转入 | 转入后余额 | 外部现金 | |---|---:|---:|---:|---:|---:|---:| | 起点 | 75.00 | — | — | 8,000初始划入 | 8,000 | 4,000 | | 1 | 73.50 | −1,500 | 6,500 | 0 / 0 | 6,500 | 4,000 | | 2 | 74.20 | +700 | 7,200 | 0 / 0 | 7,200 | 4,000 | | 3 | 71.80 | −2,400 | 4,800 | 3,200 / 3,200 | 8,000 | 800 | | 4 | 75.60 | +3,800 | 11,800 | 0 / 0 | 11,800 | 800 | 第一步恰好等于MM，不触发追缴；第三步从4,800补回8,000，需要3,200，而不是只补到6,500。最后总额为11,800+800=12,600，较初始12,000增加600。3,200的转入没有成为额外亏损。 再看路径B：75→75.2→75.4→75.5→75.6。四次VM依次为+200、+200、+100、+100，无追缴；最后担保余额8,600、外部现金4,000，总额同样为12,600。 两条路径的期末价格相同，累计VM都等于$1{,}000(75.6-75)=600$。但在路径A上，经营者有一段时间只能自由使用800美元；路径B则一直保留4,000。其他业务若也在这期间需要付款，这个差别就有实际后果。终点P&L把它省掉了，资金管理不能省。
交互里先比较A/B，再把“转款是否赶上模型截止”改为否。在首次需要补款的阶段，余额总额并未凭空消失，但未按时满足的要求仍会使路径停止。页面不把“还有钱”替换成“已按时交钱”。 ## 四、市场要求也会改变：为什么余额超过MM仍可能未补足 现在沿路径A，在第2次结算变化计入后，IM改为10,000、MM改为8,500，并从此维持。其余条件不变。这仍是教学压力变式，不是CME当时的保证金公告。 | 阶段 | VM | 转入前余额 | 当期IM/MM | 要求转入 | 实际转入 | 尚欠 | 转入后余额/外部现金 | |---|---:|---:|---|---:|---:|---:|---| | 1 | −1,500 | 6,500 | 8,000 / 6,500 | 0 | 0 | 0 | 6,500 / 4,000 | | 2，要求改变 | +700 | 7,200 | 10,000 / 8,500 | 2,800 | 2,800 | 0 | 10,000 / 1,200 | | 3 | −2,400 | 7,600 | 10,000 / 8,500 | 2,400 | 1,200 | **1,200** | 8,800 / 0；停止 | 注意最后的8,800已经高于MM=8,500，但既已触发的要求是补回IM=10,000，仍欠1,200。不能在交了一半后重新拿“已经高于维持线”抹掉剩余要求。这是为什么必须把“触发条件”和“触发后的应补金额”分开。 到停止点，累计VM为−3,200，总余额8,800正好等于12,000−3,200。后面的75.6只存在于输入路径里，不能继续画成已经获得的+3,800。若要讨论处置，需要另给真实或教学的融资、强平、费用及成交条件，不能用一个自动反弹假设填空。 ## 五、历史价格与研究材料各能告诉我们什么 把同样算法用于真实历史观测：CFTC记录的2020年5月CL结算价18.27→−37.63→10.01。假设一张多头在第一个结算价建立并一直持有，两个后续变动现金为−55,900和+47,640美元。[^MD-S03] 本篇没有该客户的账户、当时IM/MM和盘中要求，因此只重建价格对应的条件VM，不把12,000/8,000/6,500那套合成账户套成历史事实。
选读：保证金更保守，为什么也会增加资金占用？ BIS 2023年的Box A用假想利率互换对比2020美元付固定与2022英镑收固定头寸，展示VM支付和IM模型重校准可相继增加流动性需求；图中VM按五日累计，IM以五日平仓期计算。作者比较预防性压力底线与随近期波动变化的部分：更高底线可缓和压力时的追加幅度，却提高平静时期的资金占用。[^MD-S08] 读Graph A1时先找这些口径，不要把IM曲线的水平当作每日新付费用，也不要将两个互换模型当作我们的CL账户。原图提供研究模型比较，本文的路径A/B提供可亲自重算的现金机制；两者职责不同。
## 六、把“账能对上”升级为“路径能走完” **题一：边界究竟在哪里？** 路径A第一次结算后6,500，要不要补钱？若价格不是73.50而是73.49呢？ **解析。** 前者等于MM，不追缴。后者VM为−1,510，转入前6,490，严格低于MM；应补$8{,}000-6{,}490=1{,}510$。不是只补10美元。触发不等式与补款目标承担不同职责。 **题二：少一点现金。** 原始IM/MM不变，起点总现金改为11,000，路径A能否走完？ **解析。** 初始划入8,000后外部仅3,000。第三步仍需3,200，最多转入3,000，差200；账户到7,800，外部为0，停在该阶段。不能因为端点从75到75.6就报告盈利600已实现。 **题三：为何不把3,200追加再从12,600扣掉？** **解析。** 3,200来自原来4,000的外部现金。转入后它在担保余额中，已经计入11,800；外部只剩800。再次扣除相当于把自己两个账户间的移动当成对外费用。应检查“担保余额+外部现金＝初始资金+累计VM”。 **题四：加保证金情景的8,800是否合格？** **解析。** 本例在7,600低于8,500时已经产生2,400补足要求。只交1,200后虽然超过维持线，仍未达到这笔要求的10,000目标。既有追缴不能自行用重新测试维持线的方法撤销。 **题五：有现金但晚到。** 第三步需3,200且外部有4,000，但转款将在模型截止后到达，可以先计作已付吗？ **解析。** 不可以。已拥有现金、可在指定时间使用、已履行转款是三个状态。模型在截止时把实际转入记为0、未满足记为3,200并停止；它不预测真实券商将采取何种处置。 [^MD-S05]: CME，*Money Calculations for CME-cleared Futures and Options*，2015-06-11，pp.1–3正常期货变动现金单元。[原文](https://www.cmegroup.com/clearing/files/CME-Money-Calculations-Futures-and-Options.pdf)。本例CL价格与乘数使金额精确到美元；其他产品须按其舍入规则。 [^MD-S05-MARGIN]: CME，*The Benefits of Futures Margins*，保证金与初始/维持保证金单元。[原文](https://www.cmegroup.com/education/courses/understanding-the-benefits-of-futures/the-benefits-of-futures-margins)。具体8,000/6,500与严格不等式由本教学账户设定。 [^MD-S03]: CFTC staff，*WTI Interim Staff Report*，2020-11-23，印刷pp.8–9、11。[原文](https://www.cftc.gov/media/5296/InterimStaffReportNYMEX_WTICrudeOil/download)。结算价为历史观测，持仓与资金账户为条件算例。 [^MD-S08]: Benjamin H Cohen、Kevin Tracol，BIS Quarterly Review，March 2023，Box A，pp.5–6及Graph A1/脚注。[原文](https://www.bis.org/publications/perceptions-risk-and-policy-outlook-drive-markets_1.pdf)。假想IRS；没有复刻Clarus底层数据。 ## Experiment inputs and static equivalents ```json [ { "id": "EXP-MD-M11-MARGIN", "title": "每日结算与期货保证金：交互实验", "anchor": "m11-paths", "description": "固定教学截止下，先计VM，再比较MM，按IM补款并在不足/晚到停止。", "inputs": { "data_keys": [ "margin", "contract", "history" ], "adjustable": { "path": [ "A", "B" ], "contracts": [ 1, 50 ], "starting_cash": [ 0, 100000000 ], "IM/MM_per_contract": "nonnegative, MM<=IM", "raised": "stage2 after mark", "transfer_on_time": "boolean" }, "read_only": [ "price vectors for A/B", "raisedIM10000/MM8500", "deadline D_i definition" ] }, "algorithm": "Before=M+VM; required=(before0 STOP. Partial M>MM does not cancel prior IM call. Initial funding failure means no position.", "outputs": { "default": { "status": "completed", "inputs": { "identity": "teaching_account_not_CCP_or_historical_margin", "paths": { "A": [ 75, 73.5, 74.2, 71.8, 75.6 ], "B": [ 75, 75.2, 75.4, 75.5, 75.6 ] }, "contracts": 1, "initial_per_contract": 8000, "maintenance_per_contract": 6500, "starting_cash": 12000, "raised_initial_per_contract": 10000, "raised_maintenance_per_contract": 8500, "change_at_mark": 2, "deadline": "model stage D_i after settlement, before next mark; not a real broker deadline", "transfer_on_time": true, "path": "A", "raised": false }, "initial_state": { "stage": 0, "price": 75, "balance": 8000, "free": 4000, "equity": 12000 }, "rows": [ { "stage": 1, "price": 73.5, "vm": -1500, "initial": 8000, "maintenance": 6500, "before": 6500, "required": 0, "paid": 0, "shortfall": 0, "balance": 6500, "free": 4000, "equity": 10500, "deadline": "D_1", "transfer_on_time": true }, { "stage": 2, "price": 74.2, "vm": 700, "initial": 8000, "maintenance": 6500, "before": 7200, "required": 0, "paid": 0, "shortfall": 0, "balance": 7200, "free": 4000, "equity": 11200, "deadline": "D_2", "transfer_on_time": true }, { "stage": 3, "price": 71.8, "vm": -2400, "initial": 8000, "maintenance": 6500, "before": 4800, "required": 3200, "paid": 3200, "shortfall": 0, "balance": 8000, "free": 800, "equity": 8800, "deadline": "D_3", "transfer_on_time": true }, { "stage": 4, "price": 75.6, "vm": 3800, "initial": 8000, "maintenance": 6500, "before": 11800, "required": 0, "paid": 0, "shortfall": 0, "balance": 11800, "free": 800, "equity": 12600, "deadline": "D_4", "transfer_on_time": true } ], "cumulative_vm": 600, "equity_at_stop": 12600, "completed_equity": 12600, "unmodeled_marks": 0, "actual_account": false }, "B": { "status": "completed", "inputs": { "identity": "teaching_account_not_CCP_or_historical_margin", "paths": { "A": [ 75, 73.5, 74.2, 71.8, 75.6 ], "B": [ 75, 75.2, 75.4, 75.5, 75.6 ] }, "contracts": 1, "initial_per_contract": 8000, "maintenance_per_contract": 6500, "starting_cash": 12000, "raised_initial_per_contract": 10000, "raised_maintenance_per_contract": 8500, "change_at_mark": 2, "deadline": "model stage D_i after settlement, before next mark; not a real broker deadline", "transfer_on_time": true, "path": "B", "raised": false }, "initial_state": { "stage": 0, "price": 75, "balance": 8000, "free": 4000, "equity": 12000 }, "rows": [ { "stage": 1, "price": 75.2, "vm": 200, "initial": 8000, "maintenance": 6500, "before": 8200, "required": 0, "paid": 0, "shortfall": 0, "balance": 8200, "free": 4000, "equity": 12200, "deadline": "D_1", "transfer_on_time": true }, { "stage": 2, "price": 75.4, "vm": 200, "initial": 8000, "maintenance": 6500, "before": 8400, "required": 0, "paid": 0, "shortfall": 0, "balance": 8400, "free": 4000, "equity": 12400, "deadline": "D_2", "transfer_on_time": true }, { "stage": 3, "price": 75.5, "vm": 100, "initial": 8000, "maintenance": 6500, "before": 8500, "required": 0, "paid": 0, "shortfall": 0, "balance": 8500, "free": 4000, "equity": 12500, "deadline": "D_3", "transfer_on_time": true }, { "stage": 4, "price": 75.6, "vm": 100, "initial": 8000, "maintenance": 6500, "before": 8600, "required": 0, "paid": 0, "shortfall": 0, "balance": 8600, "free": 4000, "equity": 12600, "deadline": "D_4", "transfer_on_time": true } ], "cumulative_vm": 600, "equity_at_stop": 12600, "completed_equity": 12600, "unmodeled_marks": 0, "actual_account": false }, "raised": { "status": "stopped_at_unmet_call", "inputs": { "identity": "teaching_account_not_CCP_or_historical_margin", "paths": { "A": [ 75, 73.5, 74.2, 71.8, 75.6 ], "B": [ 75, 75.2, 75.4, 75.5, 75.6 ] }, "contracts": 1, "initial_per_contract": 8000, "maintenance_per_contract": 6500, "starting_cash": 12000, "raised_initial_per_contract": 10000, "raised_maintenance_per_contract": 8500, "change_at_mark": 2, "deadline": "model stage D_i after settlement, before next mark; not a real broker deadline", "transfer_on_time": true, "path": "A", "raised": true }, "initial_state": { "stage": 0, "price": 75, "balance": 8000, "free": 4000, "equity": 12000 }, "rows": [ { "stage": 1, "price": 73.5, "vm": -1500, "initial": 8000, "maintenance": 6500, "before": 6500, "required": 0, "paid": 0, "shortfall": 0, "balance": 6500, "free": 4000, "equity": 10500, "deadline": "D_1", "transfer_on_time": true }, { "stage": 2, "price": 74.2, "vm": 700, "initial": 10000, "maintenance": 8500, "before": 7200, "required": 2800, "paid": 2800, "shortfall": 0, "balance": 10000, "free": 1200, "equity": 11200, "deadline": "D_2", "transfer_on_time": true }, { "stage": 3, "price": 71.8, "vm": -2400, "initial": 10000, "maintenance": 8500, "before": 7600, "required": 2400, "paid": 1200, "shortfall": 1200, "balance": 8800, "free": 0, "equity": 8800, "deadline": "D_3", "transfer_on_time": true } ], "cumulative_vm": -3200, "equity_at_stop": 8800, "completed_equity": null, "unmodeled_marks": 1, "actual_account": false } }, "boundaries": [ "初始资金不足：无成交/无路径", "迟到/缺款：completed_equity=null, no later VM", "转入不重复扣总权益", "仅教学客户账，不是SPAN/April2020margin" ], "static_equivalent": "/notebook/labs/m-d/static/M11.html；reader两张完整表与五题解析" } ] ``` ## Sources - [WTI Interim Staff Report: Trading leading up to, on and around April 20, 2020](https://www.cftc.gov/media/5296/InterimStaffReportNYMEX_WTICrudeOil/download): 历史May结算与May−June价差；June由两者相减得到。报告不识别负价唯一根因；未采用April2020保证金。 - [Money Calculations for CME-cleared Futures and Options](https://www.cmegroup.com/clearing/files/CME-Money-Calculations-Futures-and-Options.pdf): 当日成交与期初头寸的结算现金及正常舍入；不采用无关分数报价例的印刷错误，不提供CL保证金水平。 本批读取范围：逐日价格变动、乘数、现金与结算口径。 - [The Benefits of Futures Margins](https://www.cmegroup.com/education/courses/understanding-the-benefits-of-futures/the-benefits-of-futures-margins): 保证金是应维持的担保余额而非货款；补足规则因制度/经纪商而异。不给百分比当作当前CL参数。 - [Market turbulence and soaring margins: lessons from two recent episodes — Box A](https://www.bis.org/publications/perceptions-risk-and-policy-outlook-drive-markets_1.pdf): 假想一年期USD付固定与GBP收固定IRS；VM五日累计/IM五日closeout，讨论模型重校准与流动性。未复现Clarus曲线，不替代AECOM CSA或CL客户账户。 本批读取范围：VM与模型IM在压力中共同变化；VM图为五日累计。 ## Content relations ```json [ { "from": "zh-m11", "relation": "part_of", "to": "markets-derivatives", "reason": "主要 topic 归属" }, { "from": "zh-m11", "relation": "requires", "to": "zh-m10", "required_competence": "会按方向、张数和合约单位算价格变化损益", "reason": "本篇实际需要的能力，非路径相邻推定" }, { "from": "zh-m11", "relation": "illustrated_by", "to": "EXP-MD-M11-MARGIN", "reason": "固定教学截止下，先计VM，再比较MM，按IM补款并在不足/晚到停止。" }, { "from": "zh-m11", "relation": "uses_method", "to": "zh-m04", "reason": "调用交易后担保与履约层次" }, { "from": "m11-purpose", "relation": "supported_by", "to": "MD-S05", "reason": "先按正确时段算VM，再进入资金账户", "locator": "PDF pp1–3 Normal Futures Variation", "scope": "日初持仓/当日成交的结算变化现金算法" }, { "from": "m11-accounts", "relation": "supported_by", "to": "MD-S05-MARGIN", "reason": "机制与本篇8,000/6,500教学水平分开", "locator": "两组保证金正文", "scope": "initial/maintenance及客户资金安排机制" }, { "from": "m11-history", "relation": "supported_by", "to": "MD-S03", "reason": "真实历史VM与合成IM/MM分开", "locator": "III.B–C；printed pp6–10及p11价格输入", "scope": "真实历史价格对应的条件VM" }, { "from": "m11-history", "relation": "supported_by", "to": "MD-S08", "reason": "用模型比较说明VM/IM与及时现金需求；不复刻数据、不冒充公司CSA", "locator": "Box A printed pp5–6，Graph A1/脚注", "scope": "VM/IM与压力期流动性需求的选读研究" }, { "from": "zh-m11", "relation": "illustrated_by", "to": "case-md-cl-202004", "reason": "固定时点原件案例；与教学持仓/规则版本分别记录" } ] ``` ## Related entries ## Optional reading path 理解合约与市场机制: step 9/17 重建VM、补款和资金截止，找出不能继续持有的阶段。 现金时点明确后，比较持有实物与未来取得的两条路径。 Next: [基差、持有成本与期限结构](https://ou-liu-red-sugar.github.io/zh/notebook/basis-carry-term-structure/)