使用MQL5经济日历进行交易（第一部分）：精通MQL5经济日历的功能·综合运用
（3/3）·从概览到代码再到系统整合，一篇补齐经济日历驱动交易的最后拼图
用财经日历框定品种的新闻窗口
想在 MT5 里避开重大数据行情，第一步是确认当前图表品种跟目标货币有没有关系。下面这段代码先拆出基础货币与报价货币，若两者都不含过滤货币就直接退出，避免对无关品种做无用扫描。 string currency_base = SymbolInfoString(_Symbol, SYMBOL_CURRENCY_BASE); string currency_quote = StringSubstr(_Symbol, 3, 3); if (currency_base != currency_filter && currency_quote != currency_filter){ Print("Currency (", currency_base, " | ", currency_quote, ") is not equal equal to ", currency_filter); return false; } // 第1行：取当前品种基础货币，如 EURUSD 得到 "EUR" // 第2行：从符号第4字符起截3字符，得到报价货币 "USD" // 第3-6行：若基础与报价都不是过滤项（如"USD"），打印并return false终止 时间窗口用交易日服务器时间加减一天来定。PeriodSeconds(PERIOD_D1) 返回 86400 秒，于是往前看一天、往后看一天，覆盖已发布和未发布的中等重要性事件。 datetime timeRange = PeriodSeconds(PERIOD_D1); datetime timeBefore = TimeTradeServer() - timeRange; datetime timeAfter = TimeTradeServer() + timeRange; // timeBefore 约为当前-86400秒，timeAfter 约为当前+86400秒 扫描逻辑只统计 CALENDAR_IMPORTANCE_MODERATE 级以上且落在窗口内的新闻。若 values[i].time 小于等于服务器时间且大于等于 timeBefore，就是已发布；反之在 timeAfter 内就是未发布，两者都让 totalNews 自增。 if (StringFind(_Symbol, country.currency) >= 0){ if (event.importance == CALENDAR_IMPORTANCE_MODERATE){ if (values[i].time <= TimeTradeServer() && values[i].time >= timeBefore){ Print(event.name, " > ", country.currency, " > ", EnumToString(event.importance), " Time= ", values[i].time, " (ALREADY RELEASED)"); totalNews++; } if (values[i].time >= TimeTradeServer() && values[i].time <= timeAfter){ Print(event.name, " > ", country.currency, " > ", EnumToString(event.importance), " Time= ", values[i].time, " (NOT YET RELEASED)"); totalNews++; } } } // 第1行：品种名里能找到该国货币才继续 // 第2行：只处理中等重要性事件 // 第3-6行：已发布分支，打印并计数 // 第7-10行：未发布分支，打印并计数 最终靠 totalNews 是否大于 0 决定 isNews 布尔值。外汇与贵金属受数据冲击跳空概率高，实盘前建议把这段塞进 EA 的 OnTick 开头做开关，验证时可以先把 currency_filter 写死 "USD" 看 XAUUSD 是否被正确拦截。
class="type">class="kw">string currency_base = SymbolInfoString(_Symbol, SYMBOL_CURRENCY_BASE); class="type">class="kw">string currency_quote = StringSubstr(_Symbol, class="num">3, class="num">3); if (currency_base != currency_filter && currency_quote != currency_filter){ Print("Currency(", currency_base, " | ", currency_quote, ") is not equal equal to ", currency_filter); class="kw">return class="kw">false; } if (StringFind(_Symbol, country.currency) >= class="num">0){ if (event.importance == CALENDAR_IMPORTANCE_MODERATE){ if (values[i].time <= TimeTradeServer() && values[i].time >= timeBefore){ Print(event.name, " > ", country.currency, " > ", EnumToString(event.importance), " Time= ", values[i].time, " (ALREADY RELEASED)"); totalNews++; } if (values[i].time >= TimeTradeServer() && values[i].time <= timeAfter){ Print(event.name, " > ", country.currency, " > ", EnumToString(event.importance), " Time= ", values[i].time, " (NOT YET RELEASED)"); totalNews++; } } } class="type">class="kw">datetime timeRange = PeriodSeconds(PERIOD_D1); class="type">class="kw">datetime timeBefore = TimeTradeServer() - timeRange; class="type">class="kw">datetime timeAfter = TimeTradeServer() + timeRange; Print("FURTHEST TIME LOOK BACK = ", timeBefore, " >>> CURRENT = ", TimeTradeServer()); if (totalNews > class="num">0){ isNews = true; Print(">>>>>>> (FOUND NEWS) TOTAL NEWS = ", totalNews, "/", ArraySize(values)); } else if (totalNews <= class="num">0){ isNews = class="kw">false; Print(">>>>>>> (NOT FOUND NEWS) TOTAL NEWS = ", totalNews, "/", ArraySize(values)); } class="type">class="kw">datetime startTime = TimeTradeServer() - PeriodSeconds(PERIOD_D1); class="type">class="kw">datetime endTime = TimeTradeServer() + PeriodSeconds(PERIOD_D1); class=class="str">"cmt">//+------------------------------------------------------------------+ class=class="str">"cmt">//| FUNCTION TO GET NEWS EVENTS | class=class="str">"cmt">//+------------------------------------------------------------------+ class="type">bool isNewsEvent(){ class="type">int totalNews = class="num">0; class="type">bool isNews = class="kw">false; MqlCalendarValue values[]; class="type">class="kw">datetime startTime = TimeTradeServer() - PeriodSeconds(PERIOD_D1); class="type">class="kw">datetime endTime = TimeTradeServer() + PeriodSeconds(PERIOD_D1); class=class="str">"cmt">//class="type">class="kw">string currency_filter = "USD"; class=class="str">"cmt">//class="type">class="kw">string currency_base = SymbolInfoString(_Symbol, SYMBOL_CURRENCY_BASE); class=class="str">"cmt">//class="type">class="kw">string currency_quote = StringSubstr(_Symbol, class="num">3, class="num">3); class=class="str">"cmt">//if (currency_base != currency_filter && currency_quote != currency_filter){ }
「用财经日历筛出当前品种的中等影响事件」
这段逻辑干的事很直接：拉取一段区间内的财经日历数据，只挑出和当前图表货币对相关、且重要性为中等（CALENDAR_IMPORTANCE_MODERATE）的事件，再按发布时间落在「往前一天」到「往后一天」范围内做计数。 核心调用是 CalendarValueHistory，把返回条数存进 valuesTotal，同时用 ArraySize(values) 核对实际数组容量；若两者不一致，说明缓冲区可能不够，需要调大 values 数组长度再跑。 timeBefore 与 timeAfter 以 TimeTradeServer() 加减 PERIOD_D1 的秒数得到，也就是以交易服务器时间为锚，前后各扩一天。循环里先用 StringFind(_Symbol, country.currency) 做币种匹配，再用 values[i].time 判断已发布或未发布，命中就给 totalNews 加一。 最后靠 totalNews 是否大于 0 把 isNews 置真或假，并在日志打出「FOUND NEWS / NOT FOUND NEWS」及命中占比。外汇与贵金属受新闻扰动大、滑点风险高，这类过滤只能提示概率性波动窗口，不能当作方向依据；开 MT5 把这段塞进 EA 的 NewsFilter 函数里，改一下 importance 阈值就能看不同级别事件的实际命中数。
class=class="str">"cmt">// Print("Currency(",currency_base," | ",currency_quote, class=class="str">"cmt">// ") is not equal equal to ",currency_filter); class=class="str">"cmt">// class="kw">return class="kw">false; class=class="str">"cmt">//} class="type">int valuesTotal = CalendarValueHistory(values,startTime,endTime,NULL,NULL); Print("TOTAL VALUES = ",valuesTotal," || Array size = ",ArraySize(values)); class=class="str">"cmt">//if (valuesTotal >=class="num">0 ){ class=class="str">"cmt">// Print("Calendar values as they are: "); class=class="str">"cmt">// ArrayPrint(values); class=class="str">"cmt">//} class="type">class="kw">datetime timeRange = PeriodSeconds(PERIOD_D1); class="type">class="kw">datetime timeBefore = TimeTradeServer() - timeRange; class="type">class="kw">datetime timeAfter = TimeTradeServer() + timeRange; Print("Current time = ",TimeTradeServer()); Print("FURTHEST TIME LOOK BACK = ",timeBefore," >>> LOOK FORE = ",timeAfter); for (class="type">int i = class="num">0; i < valuesTotal; i++){ MqlCalendarEvent event; CalendarEventById(values[i].event_id,event); MqlCalendarCountry country; CalendarCountryById(event.country_id,country); if (StringFind(_Symbol,country.currency) >= class="num">0){ if (event.importance == CALENDAR_IMPORTANCE_MODERATE){ if (values[i].time <= TimeTradeServer() && values[i].time >= timeBefore){ Print(event.name," > ", country.currency," > ", EnumToString(event.importance)," Time= ",values[i].time," (ALREADY RELEASED)"); totalNews++; } if (values[i].time >= TimeTradeServer() && values[i].time <= timeAfter){ Print(event.name," > ", country.currency," > ", EnumToString(event.importance)," Time= ",values[i].time," (NOT YET RELEASED)"); totalNews++; } } } } if (totalNews > class="num">0){ isNews = true; Print(">>>>>>> (FOUND NEWS) TOTAL NEWS = ",totalNews,"/",ArraySize(values)); } else if (totalNews <= class="num">0){ isNews = class="kw">false; Print(">>>>>>> (NOT FOUND NEWS) TOTAL NEWS = ",totalNews,"/",ArraySize(values)); } class="kw">return (isNews); }
◍ 把日历信号接进图表和下单逻辑
前面两节把经济日历的抓取、按货币与事件重要性过滤、区分待发生与已发生做成了结构化流程。这套流程的核心价值不在于看新闻，而在于让 EA 能在高影响事件前后自动调整行为，外汇与贵金属市场对此类事件的反应往往伴随瞬时跳空与滑点，属于典型高风险场景。 下一阶段的直接落点是把过滤后的数据画到图表窗口，并让 EA 能依据重大事件主动开仓。也就是说，系统会从「读日历」跨到「用日历下单」，设计上要以实时框架为基准，在事件触发后而非预测前去动作。 想验证这条路径，可先打开 MT5 把本系列附带的 MQL5_NEWS_CALENDAR_PART_1.mq5 跑起来，观察其过滤输出是否与你手动核对的财经日历一致；不一致的地方就是后续要改的事件权重参数。