华泰评注册制改革提速：利好大金融 首推两类股_网易财经
华泰评注册制改革提速：利好大金融 首推两类股
用微信扫码二维码
分享至好友和朋友圈
（原标题：华泰评注册制改革提速：利好大金融 首推两类股）
注册制法规呼之欲出，制度改革再提速国务院总理李克强12月9日主持召开国务院常务会议，会议通过《关于授权国务院在实施股票发行注册制改革中调整适用〈中华人民共和国证券法〉有关规定的决定（草案）》。草案明确，在决定施行之日起两年内，授权对拟在上海证券交易所、深圳证券交易所上市交易的股票公开发行实行注册制度。【点评】注册制是资本市场融资功能回归的关键，是赋予市场自我优化的核心。新规为注册制提速，三个方面彰显改革决心。首先，草案明确在决定施行之日起两年内，授权对拟在上海证券交易所、深圳证券交易所上市交易的股票公开发行实行注册制度。坚持推进股票发行注册制方向不变，授权发行时间表初步框定；其次，调整适用《证券法》关于公开发行制度的规定，制定注册制相关法规并向社会公开征求意见后公布实施，意味着股票发行注册制将先于《证券法》推出，加速落实凸显决策层实践制度改革的力度；最后，注册制继续遵循平稳过渡原则，不会一步到位，而是坚持改革力度、节奏和市场接受程度的统一，以防新股大幅扩容造成市场非理性波动。注册制将重塑券商竞争力。作为制度性的改革，注册制将彻底改变原有券商投行功能，将真正赋予投资银行市场定价、风险控制和销售能力，不仅给投行带来直接业务量的提升，更重要的是带来投行及相关业务的革命性变化。注册制推动券商全产业链转型。注册制下对投行承揽项目和资本市场部销售项目的能力、管理平台的效率都有更高要求，企业批量上市成为可能。同时将延伸企业的再融资、债券融资、并购业务需求，进一步打开券商产业链上下游的市场空间。🔥近期保险资金大举进场，昭示注册制动作加快，实际上是嗅到了新股低成本发行的溢价诱惑。证监会现阶段坚持新股低价持续发行原则，既保持投资者认购热情，也能为新股中签者带来稳妥盈利，为注册制落地酝酿积极情绪。在这个过程中，优秀大金融的高股息回报和长期稳定收益将是最大亮点。长线资金全仓优秀大金融，用两融放大杠杆，明年重点参与市值配售，年化收益能达15%以上，加上大金融3%以上股息收益率（免税），不失为上佳策略。推荐投行布局领先的券商股和创投股。
标准一旦建立了之后，大多数时间并不用总管着市场。
巨丰投顾6月策略：市场低谷期防御为上。
此种战法是短线交易者必须注意的一个因素，是散户能够获利的力量之源。
K线是分析和判断行情走势最基本的技术指标。学会看图，赚钱不再难！
在这个市场里，主力决定着大市走向。散户能赚钱的方法就是跟庄操作。
要做股票，先看大盘，如果把握不好节奏，想在市场生存，真的很难。
本文来源：华泰证券
责任编辑：曹帅_NF5384
用微信扫码二维码
分享至好友和朋友圈
加载更多新闻
热门产品:　　　
:　　　　　　　　
:　　　　　　　　　
热门影院：
阅读下一篇
用微信扫描二维码
分享至好友和朋友圈&figure&&img src=&https://pic3.zhimg.com/v2-5c916d3c21da2d18c951f8_b.jpg& data-rawwidth=&1920& data-rawheight=&1080& class=&origin_image zh-lightbox-thumb& width=&1920& data-original=&https://pic3.zhimg.com/v2-5c916d3c21da2d18c951f8_r.jpg&&&/figure&&p&本系列由&b&&a href=&http://link.zhihu.com/?target=https%3A//www.digquant.com.cn/club.php& class=& wrap external& target=&_blank& rel=&nofollow noreferrer&&点宽DigQuant量化人才俱乐部&/a&&/b&独家发布&/p&&p&未经允许，不得转载&/p&&figure&&img src=&http://pic3.zhimg.com/v2-908ade08a23f39d1d4ee_b.jpg& data-caption=&& data-rawwidth=&1400& data-rawheight=&462& class=&origin_image zh-lightbox-thumb& width=&1400& data-original=&http://pic3.zhimg.com/v2-908ade08a23f39d1d4ee_r.jpg&&&/figure&&p&&b&原文传送门：&/b&&a href=&http://link.zhihu.com/?target=https%3A//www.digquant.com.cn/forum.php%3Fmod%3Dviewthread%26tid%3D295%26extra%3Dpage%253D1& class=& wrap external& target=&_blank& rel=&nofollow noreferrer&&【研报复现】华泰证券——人工智能选股之随机森林模型&/a& &/p&&h2&&b&一、随机森林模型简介&/b&&/h2&&p&随机森林（Random Forest）作为一种比较新的机器学习方法，近年来在业界内的关注度与受欢迎程度得到逐步提升。经典的机器学习模型是神经网络，其预测精确，但计算量也大。上世纪80年代，决策树算法首次出现，通过反复二分数据进行分类或回归，计算量大大降低。2001年Breiman在此基础上提出一种新的算法，这一算法由多个随机子集生成决策树的实例组成，故我们将其形象地称为“随机森林”。下面我们从决策树入手，一起逐步探讨随机森林的奥秘。&/p&&p&&br&&/p&&p&&b&1.1 决策树&/b&&/p&&p&日常生活中，我们对于事物的认知都是基于特征的判断与分类，例如通过胎生与否判断是否哺乳动物，根据肚脐尖圆来挑选螃蟹公母。决策树就是采用这样的思想，基于多个特征进行分类决策。在树的每个结点处，根据特征的表现通过某种规则分裂出下一层的叶子节点，终端的叶子节点即为最终的分类结果。决策树学习的关键是选择最优划分属性。随着逐层划分，决策树分支结点所包含的样本类别会逐渐趋于一致，即节点分裂时要使得节点分裂后的信息增益（Information Gain）最大。&/p&&p&&br&&/p&&p&&b&1.2 CART算法&/b&&/p&&p&目前主流的决策树算法包括C4.5和CART：C4.5每个节点可分裂为多个子节点，不支持特征的组合，只能用于分类问题：CART每个节点只分裂成两个节点，支持特征的组合，可用于分类和回归问题。在随机森林中，通常采用CART算法来选择划分属性，并使用“基尼指数”（Gini Index）来定义信息增益程度。分类问题中，假设有K个类，样本集D中的点属于第k类的概率为Pk，则其Gini指数为&/p&&figure&&img src=&http://pic3.zhimg.com/v2-462bbceebf98dc53d015a_b.jpg& data-caption=&& data-rawwidth=&1300& data-rawheight=&162& class=&origin_image zh-lightbox-thumb& width=&1300& data-original=&http://pic3.zhimg.com/v2-462bbceebf98dc53d015a_r.jpg&&&/figure&&p&Gini(D)反映了从数据集D中随机抽取两个样本，其类别标记不一致的概率，Gini(D)越小，数据集D的纯度越高。二分类问题中，若对于给定的样本集合D（|D|表示集合元素个数），根据特征A分裂为D1和D2两个不想交的部分，则分裂后的&/p&&figure&&img src=&http://pic3.zhimg.com/v2-4c48d3bf2db6_b.jpg& data-caption=&& data-rawwidth=&1501& data-rawheight=&229& class=&origin_image zh-lightbox-thumb& width=&1501& data-original=&http://pic3.zhimg.com/v2-4c48d3bf2db6_r.jpg&&&/figure&&p&节点开始，递归地在每个节点分裂时选取Gini(D,A)最小的特征A为划分属性，将训练集依特征分配到两个子节点中取，照此逐层划分，直到满足停止条件（例如信息增益小于预设阈值或达到最大层次数等）。即生成一颗可以进行分类预测的决策树。&/p&&p&&br&&/p&&p&&b&1.3 树剪枝&/b&&/p&&p&我们利用训练样本来生成决策树时，在不加限制的条件下会递归地选取特征进行二次分类直到不能继续下去为止。诚然这样产生的树对训练集数据的分类很准确，，模拟分类效果如下图所示。但不难发现对于个别点的分类区域划分有些“牵强附会”，即出现过拟合现象，造成的结果是，分类器对于样本外数据的分类效果会明显劣于样本内数据，换而言之，模型预测结果的可靠性存疑。为此，我们希望通过降低决策树的复杂程度来减少过拟合的风险，故可以主动地去掉一些分支进行简化，这一过程称为剪枝（Pruning）。&/p&&figure&&img src=&http://pic2.zhimg.com/v2-f6d68ece10b65_b.jpg& data-caption=&& data-rawwidth=&1829& data-rawheight=&939& class=&origin_image zh-lightbox-thumb& width=&1829& data-original=&http://pic2.zhimg.com/v2-f6d68ece10b65_r.jpg&&&/figure&&p&决策树的剪枝有两种思路：预剪枝（Pre-Pruning）和后剪枝（Post-Pruning）。预剪枝是在构造决策树的同时进行剪枝。每个节点在划分前需进行估计，若当前节点的划分不能带来决策树泛化性能的提升，则停止划分并将当前节点标记为叶节点。后剪枝即从已生成的决策树上裁减掉一些子树或叶节点，并将其父节点作为新的叶节点。从而简化分类树模型&/p&&p&&br&&/p&&p&&b&1.3.1预剪枝&/b&&/p&&p&在预剪枝过程中，常用的判断停止树生长方法包括以下几种：&/p&&p&1、达到最大树深度（Maximum Tree Depth），如下图中设置max_depth=3；&/p&&p&2、设置最优划分下内部/叶节点的最小样本数：小于阈值即停止子树的划分或对叶节点进行剪枝；&/p&&p&3、（在特征为离散状态下）到达此节点的样本具有相同的特征（不必一定属于同一类）；&/p&&p&4、计算每次生长对系统性能的增益，如果这个增益值小于某个阈值则不进行生长。&/p&&figure&&img src=&http://pic3.zhimg.com/v2-b253f814db5a96f614fe6d57b7ae3cce_b.jpg& data-caption=&& data-rawwidth=&1565& data-rawheight=&804& class=&origin_image zh-lightbox-thumb& width=&1565& data-original=&http://pic3.zhimg.com/v2-b253f814db5a96f614fe6d57b7ae3cce_r.jpg&&&/figure&&p&&b&1.3.2后剪枝&/b&&/p&&p&后剪枝（Post-Pruning）的剪枝过程是在决策树构造完成后删除一些子树，往往是递归地自上而下进行。后剪枝常见的算法包括：错误率降低剪枝、悲观剪枝、代价复杂度剪枝、基于错误的剪枝等。我们以代价复杂度剪枝为例来进行说明。&/p&&p&设树T的叶节点个数为|T|，t是树T的叶节点，该叶节点有Nt个样本点，其中k类的样本点有Ntk个，k=1,2,…,K，Ht(T)为叶节点t上的经验熵，alpha&0为参数，则决策树学习的损失函数可以定义为：&/p&&figure&&img src=&http://pic4.zhimg.com/v2-a9c8cf77bb7a9d7e4b26b_b.jpg& data-caption=&& data-rawwidth=&928& data-rawheight=&146& class=&origin_image zh-lightbox-thumb& width=&928& data-original=&http://pic4.zhimg.com/v2-a9c8cf77bb7a9d7e4b26b_r.jpg&&&/figure&&p&其中，经验熵为&/p&&figure&&img src=&http://pic4.zhimg.com/v2-331b7cc6ecbb0cf6d43ea3_b.jpg& data-caption=&& data-rawwidth=&845& data-rawheight=&202& class=&origin_image zh-lightbox-thumb& width=&845& data-original=&http://pic4.zhimg.com/v2-331b7cc6ecbb0cf6d43ea3_r.jpg&&&/figure&&figure&&img src=&http://pic4.zhimg.com/v2-593d24a5d0dcf0b3b04ec0f373d3219f_b.jpg& data-caption=&& data-rawwidth=&1291& data-rawheight=&191& class=&origin_image zh-lightbox-thumb& width=&1291& data-original=&http://pic4.zhimg.com/v2-593d24a5d0dcf0b3b04ec0f373d3219f_r.jpg&&&/figure&&p&这时有&/p&&figure&&img src=&http://pic4.zhimg.com/v2-aaef1d92babbfea31fea89b3e192c5a3_b.jpg& data-caption=&& data-rawwidth=&682& data-rawheight=&125& class=&origin_image zh-lightbox-thumb& width=&682& data-original=&http://pic4.zhimg.com/v2-aaef1d92babbfea31fea89b3e192c5a3_r.jpg&&&/figure&&p&其中C(T)表示模型对训练数据的预测误差，即模型与训练数据的拟合程度，|T|表示模型的复杂度，参数alpha控制两者之间的关系。简而言之，后剪枝就是在给定alpha的条件下选择损失函数最小的子树。损失函数刻画了训练集拟合程度与模型复杂度之间的平衡，通过优化损失函数在进行更好拟合的同时考虑了减小模型复杂度。&/p&&p&&br&&/p&&p&&b&1.4 随机森林&/b&&/p&&p&通过前面的介绍我们已经对决策树有了清晰的了解，随机森林（RandomForest）正是一种由诸多决策树通过Bagging的方式组成的分类器。其中，Bagging是分类器集成学习的两大渊薮中区别于Boosting派系（串行方法）的一种并行方法，它的特点是各个弱学习器之间没有依赖关系，可以并行拟合。Bagging方法是Bootstrap随机采样思想在机器学习上的应用。如下图所示，我们由原始数据集生成N个Bootstrap数据集，对于每个Bootstrap数据集分别训练一个弱分类器，最终用投票、取平均值等方法组合成强分类器。&/p&&figure&&img src=&http://pic1.zhimg.com/v2-a2f2bd34_b.jpg& data-caption=&& data-rawwidth=&1582& data-rawheight=&680& class=&origin_image zh-lightbox-thumb& width=&1582& data-original=&http://pic1.zhimg.com/v2-a2f2bd34_r.jpg&&&/figure&&p&具体地说，随机森林根据以下两步方法建造每棵决策树。第一步称为“行采样”，从全体训练样本中有放回地抽样，得到一个Bootstrap数据集。第二步称为“列采样”，从全部M个特征中随机选择m个特征（m&M），以Bootstrap数据集的m个特征为新的训练集训练一棵决策树。如果是分类预测，则N棵决策树投出最多票数的类别或者类别之一为最终类别。如果是对连续数值回归预测，则对N棵决策树得到的回归或0-1分类结果进行算术平均得到的值为最终的模型输出。&/p&&p&&br&&/p&&h2&&b&二、数据说明&/b&&/h2&&p&1、特征选取：由于AT中暂时无法获取包括财务指标在内的许多基本面数据，我们这里的特征以技术面指标为主。具体包括：&/p&&p&（1）动量反转因子：包括股票过去1-6周里的对数收益率。&/p&&p&（2）波动率因子：包括股票过去1-6周里的收益率标准差与沪深300指数同期标准差的差值，大于0表明该股票在此期间的波动超过大盘，小于0表示波动小于大盘。&/p&&p&（3）资金流量因子：包括股票过去1-6周里的累计资金净流量。由于不同上市公司的规模相差很大，为了使不同股票的资金流量具有可比性，我们需要对其进行标准化处理。&/p&&p&（4）技术指标因子：取常用的技术指标MACD（长周期26天，短周期12天，DEA周期9天）和RSI（周期为20天）。&/p&&p&以上总计为20个特征指标。并以股票的超额收益率（相对于沪深300指数）作为分类指标。&/p&&p&&br&&/p&&p&2、以沪深300指数的成分股为候选股票池，并剔除测试期间未上市的以及期间累计停牌超过30天的股票。以沪深300指数成分股作为候选股票池的好处在于其中多数为业绩良好的蓝筹股，短时间里发生破产、退市等的可能性较小，因此在使用历史数据时可以减少幸存者偏差。但缺点是公司类型较为单一，训练样本的数据也较为有限。&/p&&p&&br&&/p&&p&3、训练样本期间为年。&/p&&p&&br&&/p&&p&4、投资方式：每次选股的投资期为1个月，之后进行平仓并重新选股投资。&/p&&p&（注：为便于数据处理，我们以连续的5个交易日视为1周，20个交易日视为1个月。）&/p&&p&&br&&/p&&h2&&b&三、模型训练&/b&&/h2&&p&以当前的沪深300成分股为股票池，扣除13年尚未上市以及13-15年期间累计停牌超过30天的股票，数据规模为727个交易日×155只股票，经整理得到5270组样本。对原始数据进行0-1特征分类时，上述特征指标的阈值分别取0（动量因子）、0（波动率因子）、0（资金流因子）、0（MACD）、50（RSI）和0（目标类别）。&/p&&p&我们在全体样本中随机取90%的样本作为训练集，剩下的10%样本作为验证模型泛化能力的测试集。森林规模为200棵树，特征标签的数量我们从5-20进行讨论。对于剪枝方面的考虑，我们采取前文预剪枝中的方法4，即设定信息增益阈值，当分类带来的信息增益过小时便不再进行细分。&/p&&figure&&img src=&http://pic4.zhimg.com/v2-f7b73de51f6f2de553b9c6de65c12953_b.jpg& data-rawwidth=&509& data-rawheight=&415& class=&origin_image zh-lightbox-thumb& width=&509& data-original=&http://pic4.zhimg.com/v2-f7b73de51f6f2de553b9c6de65c12953_r.jpg&&&figcaption& 注：本表格的数据是在信息增益阈值取0.0001的条件下取得的。&/figcaption&&/figure&&p&关于上表指标的说明：&/p&&p&我们记tp为真正例，即实际上涨并且模型判定为上涨的样本；fp为伪正例，即实际下跌但模型判定为上涨的样本；tn为真负例，即实际下跌且模型判定为下跌的样本；fn为伪负例，即实际上涨但模型判定为下跌的样本。&/p&&p&则准确率=(tp+tn)/(tp+fp+tn+fn)表示模型做出一个判断时，该判断的的正确率；&/p&&p&精确率=tp/(tp+fp)表示模型判断为上涨的样本中实际也是上涨的比率；&/p&&p&召回率=tp/(tp+fn)表示在整个市场的所有上涨机会中，模型能够抓住的机会的比率。&/p&&p&&br&&/p&&p&上表的数据大体上反映出了几个情况：&/p&&p&（1）对于训练集数据，随着特征数量的增加，准确率和召回率都呈明显的上涨趋势，而精确率基本都超过0.8这一相当高的数据，但对于测试集数据，各项指标基本上都明显小于训练集数据。表明模型依然存在过拟合现象。&/p&&p&&br&&/p&&p&（2）我们的评估应以测试集的指标为主，其中精确率反映了投资标的（预测上涨）中真正上涨的比率，因此是我们主要的决策指标，同时召回率不应过低，否则当市场机会较少时可能会出现找不出投资标的的情况（即模型全部判定为下跌）。综合这两个指标，我们认为特征数量应在10-15区间为宜。&/p&&p&&br&&/p&&p&（3）由于模型本身是一种随机算法，因此即使在相同的参数下，每次计算的结果也会有所出入，为了进一步确保模型的可靠性，我们还应多次计算并考察指标的分布情况&/p&&p&关于过拟合，我们调整了信息增益阈值的参数，但依然不能很好地消除该问题。由于时间关系，我们暂时不讨论其他剪枝方法下是否能更好地解决过拟合的问题。&/p&&p&&br&&/p&&p&我们选择上表中指标较为出色的特征个数参数（10、11和14）进行进一步的验证。考虑到计算时间的问题，每一个参数我们都只进行10次计算，并将每一次的计算结果如下表所示：&/p&&figure&&img src=&http://pic1.zhimg.com/v2-f8e7c21e440_b.jpg& data-rawwidth=&725& data-rawheight=&270& class=&origin_image zh-lightbox-thumb& width=&725& data-original=&http://pic1.zhimg.com/v2-f8e7c21e440_r.jpg&&&figcaption&表3.2关于模型稳定性的多次训练&/figcaption&&/figure&&p&总的来讲，当特征个数在10个左右时，模型在样本外也有较好的预测效果&/p&&p&&br&&/p&&h2&&b&四、策略回测&/b&&/h2&&p&我们用表3.2中测试集指标中精确率最高的森林网络（特征个数为10时的第六次训练结果）进行2016年的回测，回测效果如下图所示：&/p&&figure&&img src=&http://pic3.zhimg.com/v2-789abe14c950a03a215bd4b3ef62759a_b.jpg& data-caption=&& data-rawwidth=&1576& data-rawheight=&812& class=&origin_image zh-lightbox-thumb& width=&1576& data-original=&http://pic3.zhimg.com/v2-789abe14c950a03a215bd4b3ef62759a_r.jpg&&&/figure&&p&主要业绩指标：年化收益15.6%，最大回撤18.5%，夏普比率0.63，Calmar比率0.74，胜率57.41%。&/p&&p&&br&&/p&&p&总结：总的来说，随机森林算法只是提供了一种分类方法，该方法在实践中的有效性除了受算法本身的影响以外，更依赖于分类的依据（特征指标），如果选取的特征指标对事件的预测（相关）能力较弱，那么再好的算法可能也很难在样本外取得很好的分类效果。因此对于本文的进一步改进方向笔者认为一是关于所选取特征指标的考量，二是如何改进算法（例如剪枝部分）以提高模型的泛化能力。&/p&&p&&br&&/p&&p&&b&? 回测软件：&a href=&http://link.zhihu.com/?target=https%3A//www.digquant.com.cn/at.php& class=& wrap external& target=&_blank& rel=&nofollow noreferrer&&Auto-Trader Pro&/a&&/b&（AT 量能策略研究终端 以 MATLAB 研究平台为依托，甚至可以把交易思路延生到机器学习，神经网络，舆情分析等更复杂的领域。依托 Matlab 强大的数理研究及强大的金融工具箱，策略师不再需要依赖程序员的编程能力，而将研究成果直接转化为程序化策略。）&/p&&figure&&img src=&http://pic4.zhimg.com/v2-a204f22cbcb7dbc9bc37_b.jpg& data-caption=&& data-rawwidth=&660& data-rawheight=&312& class=&origin_image zh-lightbox-thumb& width=&660& data-original=&http://pic4.zhimg.com/v2-a204f22cbcb7dbc9bc37_r.jpg&&&/figure&&p&&br&&/p&&p&&b&代码部分：（基本的数据处理部分的代码从略）&/b&&/p&&p&&b&1.单棵决策树代码：&/b&&/p&&div class=&highlight&&&pre&&code class=&language-matlab&&&span&&/span&&span class=&p&&[&/span&&span class=&n&&color&/span&&span class=&p&&=&/span&&span class=&n&&black&/span&&span class=&p&&]&/span&&span class=&k&&function&/span& &span class=&n&&tree0&/span&&span class=&p&&=&/span&&span class=&n&&onetree&/span&&span class=&p&&(&/span&&span class=&n&&flabels&/span&&span class=&p&&,&/span&&span class=&n&&trainfeatures&/span&&span class=&p&&,&/span&&span class=&n&&class&/span&&span class=&p&&,&/span&&span class=&n&&epsino&/span&&span class=&p&&)&/span&
&span class=&c&&%输入变量依次为特征标签、样本的特征矩阵、样本的分类结果、信息增益阈值（关于剪枝的考量），输出为决策树（struct结构） &/span&
&span class=&n&&tree0&/span&&span class=&p&&=&/span&&span class=&n&&struct&/span&&span class=&p&&(&/span&&span class=&s&&'pro'&/span&&span class=&p&&,&/span&&span class=&mi&&0&/span&&span class=&p&&,&/span&&span class=&s&&'value'&/span&&span class=&p&&,&/span&&span class=&o&&-&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&s&&'child'&/span&&span class=&p&&,[],&/span&&span class=&s&&'parentpro'&/span&&span class=&p&&,&/span&&span class=&o&&-&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span&
&span class=&c&&%关于树结构的初始设置，pro为节点类别，1为内节点，0为叶节点；value为节点值，叶节点对应分类值，内节点对应标签值 &/span&
&span class=&c&&%child子树集，struct结构，parentpro父节点，如有则为节点标签值，若无则为-1&/span&
&span class=&n&&cn&/span&&span class=&p&&=&/span&&span class=&n&&unique&/span&&span class=&p&&(&/span&&span class=&n&&class&/span&&span class=&p&&);&/span& &span class=&n&&len&/span&&span class=&p&&=&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&cn&/span&&span class=&p&&);&/span& &span class=&c&&%类别集合、类别数 &/span&
&span class=&p&&[&/span&&span class=&n&&n&/span&&span class=&p&&,&/span&&span class=&n&&m&/span&&span class=&p&&]=&/span&&span class=&nb&&size&/span&&span class=&p&&(&/span&&span class=&n&&trainfeatures&/span&&span class=&p&&);&/span&
&span class=&k&&if&/span& &span class=&n&&len&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span& &span class=&c&&%如果只剩下一个分类，则无需再分，返回&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&pro&/span&&span class=&p&&=&/span&&span class=&mi&&0&/span&&span class=&p&&;&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&value&/span&&span class=&p&&=&/span&&span class=&n&&cn&/span&&span class=&p&&;&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&child&/span&&span class=&p&&=[];&/span&
&span class=&k&&return&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&k&&if&/span& &span class=&n&&n&/span&&span class=&o&&==&/span&&span class=&mi&&0&/span& &span class=&c&&%已无更多标签，则分类结果为数量更多的一个分类，返回&/span&
&span class=&n&&H&/span&&span class=&p&&=&/span&&span class=&n&&hist&/span&&span class=&p&&(&/span&&span class=&n&&class&/span&&span class=&p&&,&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&cn&/span&&span class=&p&&));&/span&
&span class=&p&&[&/span&&span class=&o&&~&/span&&span class=&p&&,&/span&&span class=&n&&largest&/span&&span class=&p&&]=&/span&&span class=&n&&max&/span&&span class=&p&&(&/span&&span class=&n&&H&/span&&span class=&p&&);&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&pro&/span&&span class=&p&&=&/span&&span class=&mi&&0&/span&&span class=&p&&;&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&value&/span&&span class=&p&&=&/span&&span class=&n&&cn&/span&&span class=&p&&(&/span&&span class=&n&&largest&/span&&span class=&p&&);&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&child&/span&&span class=&p&&=[];&/span&
&span class=&k&&return&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&n&&pnode&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&len&/span&&span class=&p&&);&/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&len&/span&
&span class=&n&&pnode&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)=&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&nb&&find&/span&&span class=&p&&(&/span&&span class=&n&&class&/span&&span class=&o&&==&/span&&span class=&n&&cn&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)))&/span&&span class=&o&&/&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&class&/span&&span class=&p&&);&/span&
&span class=&k&&end&/span&
&span class=&n&&Gini&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&o&&-&/span&&span class=&n&&sum&/span&&span class=&p&&(&/span&&span class=&n&&pnode&/span&&span class=&o&&.^&/span&&span class=&mi&&2&/span&&span class=&p&&);&/span&
&span class=&n&&maxium&/span&&span class=&p&&=&/span&&span class=&o&&-&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&n&&maxi&/span&&span class=&p&&=&/span&&span class=&o&&-&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&n&&g&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&n&/span&&span class=&p&&);&/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&n&/span&
&span class=&n&&fn&/span&&span class=&p&&=&/span&&span class=&n&&unique&/span&&span class=&p&&(&/span&&span class=&n&&trainfeatures&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,:));&/span& &span class=&c&&%第i个标签的所有属性集合&/span&
&span class=&n&&lfn&/span&&span class=&p&&=&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&fn&/span&&span class=&p&&);&/span&
&span class=&n&&gfn&/span&&span class=&p&&=&/span&&span class=&nb&&ones&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&lfn&/span&&span class=&p&&);&/span& &span class=&c&&%每一个属性下的gini值 &/span&
&span class=&k&&for&/span& &span class=&n&&k&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&lfn&/span&
&span class=&n&&onepro&/span&&span class=&p&&=&/span&&span class=&nb&&find&/span&&span class=&p&&(&/span&&span class=&n&&trainfeatures&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,:)&/span&&span class=&o&&==&/span&&span class=&n&&fn&/span&&span class=&p&&(&/span&&span class=&n&&k&/span&&span class=&p&&));&/span& &span class=&c&&%获取标签i下属性k的个体序号 &/span&
&span class=&k&&for&/span& &span class=&nb&&j&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&len&/span&
&span class=&n&&onepro2&/span&&span class=&p&&=&/span&&span class=&nb&&find&/span&&span class=&p&&(&/span&&span class=&n&&class&/span&&span class=&p&&(:,&/span&&span class=&n&&onepro&/span&&span class=&p&&)&/span&&span class=&o&&==&/span&&span class=&n&&cn&/span&&span class=&p&&(&/span&&span class=&nb&&j&/span&&span class=&p&&));&/span&
&span class=&n&&ratiopro&/span&&span class=&p&&=&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&onepro2&/span&&span class=&p&&)&/span&&span class=&o&&/&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&onepro&/span&&span class=&p&&);&/span&
&span class=&n&&gfn&/span&&span class=&p&&(&/span&&span class=&n&&k&/span&&span class=&p&&)=&/span&&span class=&n&&gfn&/span&&span class=&p&&(&/span&&span class=&n&&k&/span&&span class=&p&&)&/span&&span class=&o&&-&/span&&span class=&n&&ratiopro&/span&^&span class=&mi&&2&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&n&&ratio&/span&&span class=&p&&=&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&onepro&/span&&span class=&p&&)&/span&&span class=&o&&/&/span&&span class=&n&&m&/span&&span class=&p&&;&/span&
&span class=&n&&gfn&/span&&span class=&p&&(&/span&&span class=&n&&k&/span&&span class=&p&&)=&/span&&span class=&n&&gfn&/span&&span class=&p&&(&/span&&span class=&n&&k&/span&&span class=&p&&)&/span&&span class=&o&&*&/span&&span class=&n&&ratio&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&n&&g&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)=&/span&&span class=&n&&Gini&/span&&span class=&o&&-&/span&&span class=&n&&sum&/span&&span class=&p&&(&/span&&span class=&n&&gfn&/span&&span class=&p&&);&/span& &span class=&c&&%标签i带来的信息增益&/span&
&span class=&k&&if&/span& &span class=&n&&g&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)&/span&&span class=&o&&&&/span&&span class=&n&&maxium&/span&
&span class=&n&&maxium&/span&&span class=&p&&=&/span&&span class=&n&&g&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&);&/span&
&span class=&n&&maxi&/span&&span class=&p&&=&/span&&span class=&nb&&i&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&k&&if&/span& &span class=&n&&maxium&/span&&span class=&o&&&&/span&&span class=&p&&=&/span&&span class=&n&&epsino&/span& &span class=&c&&%即分类没有带来信息增益 &/span&
&span class=&n&&H&/span&&span class=&p&&=&/span&&span class=&n&&hist&/span&&span class=&p&&(&/span&&span class=&n&&class&/span&&span class=&p&&,&/span&&span class=&n&&len&/span&&span class=&p&&);&/span&
&span class=&p&&[&/span&&span class=&o&&~&/span&&span class=&p&&,&/span&&span class=&n&&largest&/span&&span class=&p&&]=&/span&&span class=&n&&max&/span&&span class=&p&&(&/span&&span class=&n&&H&/span&&span class=&p&&);&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&pro&/span&&span class=&p&&=&/span&&span class=&mi&&0&/span&&span class=&p&&;&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&value&/span&&span class=&p&&=&/span&&span class=&n&&cn&/span&&span class=&p&&(&/span&&span class=&n&&largest&/span&&span class=&p&&);&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&child&/span&&span class=&p&&=[];&/span&
&span class=&k&&return&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&pro&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&n&&tv&/span&&span class=&p&&=&/span&&span class=&n&&flabels&/span&&span class=&p&&(&/span&&span class=&n&&maxi&/span&&span class=&p&&);&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&value&/span&&span class=&p&&=&/span&&span class=&n&&tv&/span&&span class=&p&&;&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&child&/span&&span class=&p&&=[];&/span&
&span class=&n&&flabels&/span&&span class=&p&&(&/span&&span class=&n&&maxi&/span&&span class=&p&&)=[];&/span& &span class=&c&&%删除该标签&/span&
&span class=&c&&%接下来进行数据分割 &/span&
&span class=&p&&[&/span&&span class=&n&&data&/span&&span class=&p&&,&/span&&span class=&n&&class0&/span&&span class=&p&&,&/span&&span class=&n&&splitarr&/span&&span class=&p&&]=&/span&&span class=&n&&splitd&/span&&span class=&p&&(&/span&&span class=&n&&trainfeatures&/span&&span class=&p&&,&/span&&span class=&n&&class&/span&&span class=&p&&,&/span&&span class=&n&&maxi&/span&&span class=&p&&);&/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&nb&&size&/span&&span class=&p&&(&/span&&span class=&n&&data&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&)&/span&
&span class=&n&&subdata&/span&&span class=&p&&=&/span&&span class=&n&&data&/span&&span class=&p&&{&/span&&span class=&nb&&i&/span&&span class=&p&&};&/span&
&span class=&n&&temptree&/span&&span class=&p&&=&/span&&span class=&n&&onetree&/span&&span class=&p&&(&/span&&span class=&n&&flabels&/span&&span class=&p&&,&/span&&span class=&n&&subdata&/span&&span class=&p&&,&/span&&span class=&n&&class0&/span&&span class=&p&&{&/span&&span class=&nb&&i&/span&&span class=&p&&},&/span&&span class=&n&&epsino&/span&&span class=&p&&);&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&pro&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&value&/span&&span class=&p&&=&/span&&span class=&n&&tv&/span&&span class=&p&&;&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&child&/span&&span class=&p&&{&/span&&span class=&nb&&i&/span&&span class=&p&&}=&/span&&span class=&n&&temptree&/span&&span class=&p&&;&/span&
&span class=&n&&tree0&/span&&span class=&p&&.&/span&&span class=&n&&child&/span&&span class=&p&&{&/span&&span class=&nb&&i&/span&&span class=&p&&}.&/span&&span class=&n&&parentpro&/span&&span class=&p&&=&/span&&span class=&n&&splitarr&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&);&/span&
&span class=&k&&end&/span&
&span class=&k&& &/span&
&span class=&k&&function&/span&&span class=&w&& &/span&[data,class0,splitarr]&span class=&p&&=&/span&&span class=&nf&&splitd&/span&&span class=&p&&(&/span&olddata,oldclass,splitindex&span class=&p&&)&/span&&span class=&w&& &/span&
&span class=&n&&fn&/span&&span class=&p&&=&/span&&span class=&n&&unique&/span&&span class=&p&&(&/span&&span class=&n&&olddata&/span&&span class=&p&&(&/span&&span class=&n&&splitindex&/span&&span class=&p&&,:));&/span&
&span class=&n&&data&/span&&span class=&p&&=&/span&&span class=&n&&cell&/span&&span class=&p&&(&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&fn&/span&&span class=&p&&),&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span&
&span class=&n&&class0&/span&&span class=&p&&=&/span&&span class=&n&&cell&/span&&span class=&p&&(&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&fn&/span&&span class=&p&&),&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span&
&span class=&n&&splitarr&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&nb&&size&/span&&span class=&p&&(&/span&&span class=&n&&fn&/span&&span class=&p&&));&/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&fn&/span&&span class=&p&&)&/span&
&span class=&n&&fcolumn&/span&&span class=&p&&=&/span&&span class=&nb&&find&/span&&span class=&p&&(&/span&&span class=&n&&olddata&/span&&span class=&p&&(&/span&&span class=&n&&splitindex&/span&&span class=&p&&,:)&/span&&span class=&o&&==&/span&&span class=&n&&fn&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&));&/span&
&span class=&n&&data&/span&&span class=&p&&{&/span&&span class=&nb&&i&/span&&span class=&p&&}=&/span&&span class=&n&&olddata&/span&&span class=&p&&(:,&/span&&span class=&n&&fcolumn&/span&&span class=&p&&);&/span&
&span class=&n&&class0&/span&&span class=&p&&{&/span&&span class=&nb&&i&/span&&span class=&p&&}=&/span&&span class=&n&&oldclass&/span&&span class=&p&&(:,&/span&&span class=&n&&fcolumn&/span&&span class=&p&&);&/span&
&span class=&n&&data&/span&&span class=&p&&{&/span&&span class=&nb&&i&/span&&span class=&p&&}(&/span&&span class=&n&&splitindex&/span&&span class=&p&&,:)=[];&/span&
&span class=&n&&splitarr&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)=&/span&&span class=&n&&fn&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&);&/span& &span class=&c&&%用于记录分支的parent &/span&
&span class=&k&&end&/span&
&/code&&/pre&&/div&&p&&br&&/p&&p&&b&2.随机森林训练代码：&/b&&/p&&div class=&highlight&&&pre&&code class=&language-matlab&&&span&&/span&&span class=&p&&[&/span&&span class=&n&&color&/span&&span class=&p&&=&/span&&span class=&n&&black&/span&&span class=&p&&]&/span&&span class=&k&&function&/span& &span class=&p&&[&/span&&span class=&n&&rforest&/span&&span class=&p&&,&/span&&span class=&n&&trainindex&/span&&span class=&p&&,&/span&&span class=&n&&testindex&/span&&span class=&p&&]=&/span&&span class=&n&&rftrain&/span&&span class=&p&&(&/span&&span class=&n&&allsamples&/span&&span class=&p&&,&/span&&span class=&n&&epsino&/span&&span class=&p&&,&/span&&span class=&n&&ntree&/span&&span class=&p&&,&/span&&span class=&n&&nlabel&/span&&span class=&p&&)&/span&
&span class=&c&&%输入参数依次为：所有的训练样本数据，信息增益阈值，森林规模，每棵树的标签数 &/span&
&span class=&c&&%输出参数：森林网络（1*ntree的cell结构），训练样本的相关指标（准确率、精确率和召回率），测试样本的相关指标&/span&
&span class=&n&&len&/span&&span class=&p&&=&/span&&span class=&nb&&size&/span&&span class=&p&&(&/span&&span class=&n&&allsamples&/span&&span class=&p&&,&/span&&span class=&mi&&2&/span&&span class=&p&&);&/span&
&span class=&n&&sy0&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&len&/span&&span class=&p&&;&/span&
&span class=&n&&sytrain&/span&&span class=&p&&=&/span&&span class=&n&&randsample&/span&&span class=&p&&(&/span&&span class=&n&&sy0&/span&&span class=&p&&,&/span&&span class=&nb&&round&/span&&span class=&p&&(&/span&&span class=&n&&len&/span&&span class=&o&&*&/span&&span class=&mf&&0.9&/span&&span class=&p&&));&/span& &span class=&c&&%随机抽取全样本中90%的个体作为训练样本，剩余10%作为检验样本 &/span&
&span class=&n&&sytest&/span&&span class=&p&&=&/span&&span class=&n&&setdiff&/span&&span class=&p&&(&/span&&span class=&n&&sy0&/span&&span class=&p&&,&/span&&span class=&n&&sytrain&/span&&span class=&p&&);&/span&
&span class=&n&&trainsamples&/span&&span class=&p&&=&/span&&span class=&n&&allsamples&/span&&span class=&p&&(:,&/span&&span class=&n&&sytrain&/span&&span class=&p&&);&/span&
&span class=&c&&%训练样本集 &/span&
&span class=&n&&testsamples&/span&&span class=&p&&=&/span&&span class=&n&&allsamples&/span&&span class=&p&&(:,&/span&&span class=&n&&sytest&/span&&span class=&p&&);&/span&
&span class=&c&&%测试样本集&/span&
&span class=&n&&rforest&/span&&span class=&p&&=&/span&&span class=&n&&cell&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&ntree&/span&&span class=&p&&);&/span&
&span class=&p&&[&/span&&span class=&n&&m&/span&&span class=&p&&,&/span&&span class=&n&&n1&/span&&span class=&p&&]=&/span&&span class=&nb&&size&/span&&span class=&p&&(&/span&&span class=&n&&trainsamples&/span&&span class=&p&&);&/span&
&span class=&n&&trainfeatures&/span&&span class=&p&&=&/span&&span class=&n&&trainsamples&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&m&/span&&span class=&o&&-&/span&&span class=&mi&&1&/span&&span class=&p&&,:);&/span& &span class=&c&&%训练样本的特征集 &/span&
&span class=&n&&trainclass&/span&&span class=&p&&=&/span&&span class=&n&&trainsamples&/span&&span class=&p&&(&/span&&span class=&n&&m&/span&&span class=&p&&,:);&/span& &span class=&c&&%训练样本的分类集 &/span&
&span class=&n&&testfeatures&/span&&span class=&p&&=&/span&&span class=&n&&testsamples&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&m&/span&&span class=&o&&-&/span&&span class=&mi&&1&/span&&span class=&p&&,:);&/span& &span class=&c&&%测试样本特征集 &/span&
&span class=&n&&testclass&/span&&span class=&p&&=&/span&&span class=&n&&testsamples&/span&&span class=&p&&(&/span&&span class=&n&&m&/span&&span class=&p&&,:);&/span& &span class=&c&&%测试样本分类集 &/span&
&span class=&n&&flabels&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&m&/span&&span class=&o&&-&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&c&&%---------------训练------------------------------- &/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&ntree&/span&
&span class=&n&&sel1&/span&&span class=&p&&=&/span&&span class=&nb&&ceil&/span&&span class=&p&&(&/span&&span class=&nb&&rand&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&n1&/span&&span class=&p&&)&/span&&span class=&o&&*&/span&&span class=&n&&n1&/span&&span class=&p&&);&/span&
&span class=&n&&trainfeature&/span&&span class=&p&&=&/span&&span class=&n&&trainfeatures&/span&&span class=&p&&(:,&/span&&span class=&n&&sel1&/span&&span class=&p&&);&/span& &span class=&c&&%使用bootstrap方法获得每一棵树的训练样本 &/span&
&span class=&n&&class0&/span&&span class=&p&&=&/span&&span class=&n&&trainclass&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&sel1&/span&&span class=&p&&);&/span&
&span class=&n&&flabel0&/span&&span class=&p&&=&/span&&span class=&n&&randsample&/span&&span class=&p&&(&/span&&span class=&n&&flabels&/span&&span class=&p&&,&/span&&span class=&n&&nlabel&/span&&span class=&p&&);&/span&
&span class=&n&&flabel0&/span&&span class=&p&&=&/span&&span class=&n&&sort&/span&&span class=&p&&(&/span&&span class=&n&&flabel0&/span&&span class=&p&&);&/span&
&span class=&n&&trainfeature0&/span&&span class=&p&&=&/span&&span class=&n&&trainfeature&/span&&span class=&p&&(&/span&&span class=&n&&flabel0&/span&&span class=&p&&,:);&/span&
&span class=&n&&tree0&/span&&span class=&p&&=&/span&&span class=&n&&onetree&/span&&span class=&p&&(&/span&&span class=&n&&flabel0&/span&&span class=&p&&,&/span&&span class=&n&&trainfeature0&/span&&span class=&p&&,&/span&&span class=&n&&class0&/span&&span class=&p&&,&/span&&span class=&n&&epsino&/span&&span class=&p&&);&/span&
&span class=&n&&rforest&/span&&span class=&p&&{&/span&&span class=&nb&&i&/span&&span class=&p&&}=&/span&&span class=&n&&tree0&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&n&&n2&/span&&span class=&p&&=&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&testclass&/span&&span class=&p&&);&/span&
&span class=&n&&classify1&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&n1&/span&&span class=&p&&);&/span&
&span class=&n&&tc1&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&ntree&/span&&span class=&p&&);&/span&
&span class=&n&&classify2&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&n2&/span&&span class=&p&&);&/span&
&span class=&n&&tc2&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&ntree&/span&&span class=&p&&);&/span&
&span class=&c&&%----------统计训练样本的效率----------- &/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&n1&/span&
&span class=&n&&tf0&/span&&span class=&p&&=&/span&&span class=&n&&trainfeatures&/span&&span class=&p&&(:,&/span&&span class=&nb&&i&/span&&span class=&p&&);&/span&
&span class=&k&&for&/span& &span class=&nb&&j&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&ntree&/span&
&span class=&n&&tree1&/span&&span class=&p&&=&/span&&span class=&n&&rforest&/span&&span class=&p&&{&/span&&span class=&nb&&j&/span&&span class=&p&&};&/span&
&span class=&k&&while&/span& &span class=&n&&tree1&/span&&span class=&p&&.&/span&&span class=&n&&pro&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span&
&span class=&n&&childset&/span&&span class=&p&&=&/span&&span class=&n&&tree1&/span&&span class=&p&&.&/span&&span class=&n&&child&/span&&span class=&p&&;&/span&
&span class=&n&&v&/span&&span class=&p&&=&/span&&span class=&n&&tree1&/span&&span class=&p&&.&/span&&span class=&n&&value&/span&&span class=&p&&;&/span&
&span class=&n&&clen&/span&&span class=&p&&=&/span&&span class=&nb&&size&/span&&span class=&p&&(&/span&&span class=&n&&childset&/span&&span class=&p&&,&/span&&span class=&mi&&2&/span&&span class=&p&&);&/span&
&span class=&k&&if&/span& &span class=&n&&clen&/span&&span class=&o&&==&/span&&span class=&mi&&2&/span&
&span class=&k&&for&/span& &span class=&n&&k&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&clen&/span&
&span class=&n&&child&/span&&span class=&p&&=&/span&&span class=&n&&childset&/span&&span class=&p&&{&/span&&span class=&n&&k&/span&&span class=&p&&};&/span&
&span class=&k&&if&/span& &span class=&n&&child&/span&&span class=&p&&.&/span&&span class=&n&&parentpro&/span&&span class=&o&&==&/span&&span class=&n&&tf0&/span&&span class=&p&&(&/span&&span class=&n&&v&/span&&span class=&p&&)&/span&
&span class=&n&&tree1&/span&&span class=&p&&=&/span&&span class=&n&&child&/span&&span class=&p&&;&/span&
&span class=&k&&break&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&k&&else&/span&
&span class=&n&&tree1&/span&&span class=&p&&=&/span&&span class=&n&&childset&/span&&span class=&p&&{&/span&&span class=&mi&&1&/span&&span class=&p&&};&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&n&&tc1&/span&&span class=&p&&(&/span&&span class=&nb&&j&/span&&span class=&p&&)=&/span&&span class=&n&&tree1&/span&&span class=&p&&.&/span&&span class=&n&&value&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&n&&ratio&/span&&span class=&p&&=&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&nb&&find&/span&&span class=&p&&(&/span&&span class=&n&&tc1&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span&&span class=&p&&))&/span&&span class=&o&&/&/span&&span class=&n&&ntree&/span&&span class=&p&&;&/span&
&span class=&c&&%这里仅有0-1两分类，如有更多分类本句应改写 &/span&
&span class=&k&&if&/span& &span class=&n&&ratio&/span&&span class=&o&&&&/span&&span class=&mf&&0.5&/span&
&span class=&c&&%如判断为上涨的数量更多 &/span&
&span class=&n&&classify1&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)=&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&n&&tp1&/span&&span class=&p&&=&/span&&span class=&mi&&0&/span&&span class=&p&&;&/span&&span class=&n&&fp1&/span&&span class=&p&&=&/span&&span class=&mi&&0&/span&&span class=&p&&;&/span&&span class=&n&&tn1&/span&&span class=&p&&=&/span&&span class=&mi&&0&/span&&span class=&p&&;&/span&&span class=&n&&fn1&/span&&span class=&p&&=&/span&&span class=&mi&&0&/span&&span class=&p&&;&/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&n1&/span&
&span class=&k&&if&/span& &span class=&n&&classify1&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span&
&span class=&k&&if&/span& &span class=&n&&trainclass&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span&
&span class=&n&&tp1&/span&&span class=&p&&=&/span&&span class=&n&&tp1&/span&&span class=&o&&+&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&k&&else&/span&
&span class=&n&&fp1&/span&&span class=&p&&=&/span&&span class=&n&&fp1&/span&&span class=&o&&+&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&k&&else&/span&
&span class=&k&&if&/span& &span class=&n&&trainclass&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span&
&span class=&n&&fn1&/span&&span class=&p&&=&/span&&span class=&n&&fn1&/span&&span class=&o&&+&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&k&&else&/span&
&span class=&n&&tn1&/span&&span class=&p&&=&/span&&span class=&n&&tn1&/span&&span class=&o&&+&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&n&&accu1&/span&&span class=&p&&=(&/span&&span class=&n&&tp1&/span&&span class=&o&&+&/span&&span class=&n&&tn1&/span&&span class=&p&&)&/span&&span class=&o&&/&/span&&span class=&p&&(&/span&&span class=&n&&tp1&/span&&span class=&o&&+&/span&&span class=&n&&tn1&/span&&span class=&o&&+&/span&&span class=&n&&fp1&/span&&span class=&o&&+&/span&&span class=&n&&fn1&/span&&span class=&p&&);&/span&
&span class=&n&&prec1&/span&&span class=&p&&=&/span&&span class=&n&&tp1&/span&&span class=&o&&/&/span&&span class=&p&&(&/span&&span class=&n&&tp1&/span&&span class=&o&&+&/span&&span class=&n&&fp1&/span&&span class=&p&&);&/span&
&span class=&n&&reca1&/span&&span class=&p&&=&/span&&span class=&n&&tp1&/span&&span class=&o&&/&/span&&span class=&p&&(&/span&&span class=&n&&tp1&/span&&span class=&o&&+&/span&&span class=&n&&fn1&/span&&span class=&p&&);&/span&
&span class=&n&&trainindex&/span&&span class=&p&&=[&/span&&span class=&n&&accu1&/span&&span class=&p&&,&/span&&span class=&n&&prec1&/span&&span class=&p&&,&/span&&span class=&n&&reca1&/span&&span class=&p&&];&/span&
&span class=&c&&%---------统计测试样本的效率------------ &/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&n2&/span&
&span class=&n&&tf0&/span&&span class=&p&&=&/span&&span class=&n&&testfeatures&/span&&span class=&p&&(:,&/span&&span class=&nb&&i&/span&&span class=&p&&);&/span&
&span class=&k&&for&/span& &span class=&nb&&j&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&ntree&/span&
&span class=&n&&tree1&/span&&span class=&p&&=&/span&&span class=&n&&rforest&/span&&span class=&p&&{&/span&&span class=&nb&&j&/span&&span class=&p&&};&/span&
&span class=&k&&while&/span& &span class=&n&&tree1&/span&&span class=&p&&.&/span&&span class=&n&&pro&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span&
&span class=&n&&childset&/span&&span class=&p&&=&/span&&span class=&n&&tree1&/span&&span class=&p&&.&/span&&span class=&n&&child&/span&&span class=&p&&;&/span&
&span class=&n&&v&/span&&span class=&p&&=&/span&&span class=&n&&tree1&/span&&span class=&p&&.&/span&&span class=&n&&value&/span&&span class=&p&&;&/span&
&span class=&n&&clen&/span&&span class=&p&&=&/span&&span class=&nb&&size&/span&&span class=&p&&(&/span&&span class=&n&&childset&/span&&span class=&p&&,&/span&&span class=&mi&&2&/span&&span class=&p&&);&/span&
&span class=&k&&if&/span& &span class=&n&&clen&/span&&span class=&o&&==&/span&&span class=&mi&&2&/span&
&span class=&k&&for&/span& &span class=&n&&k&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&clen&/span&
&span class=&n&&child&/span&&span class=&p&&=&/span&&span class=&n&&childset&/span&&span class=&p&&{&/span&&span class=&n&&k&/span&&span class=&p&&};&/span&
&span class=&k&&if&/span& &span class=&n&&child&/span&&span class=&p&&.&/span&&span class=&n&&parentpro&/span&&span class=&o&&==&/span&&span class=&n&&tf0&/span&&span class=&p&&(&/span&&span class=&n&&v&/span&&span class=&p&&)&/span&
&span class=&n&&tree1&/span&&span class=&p&&=&/span&&span class=&n&&child&/span&&span class=&p&&;&/span&
&span class=&k&&break&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&k&&else&/span&
&span class=&n&&tree1&/span&&span class=&p&&=&/span&&span class=&n&&childset&/span&&span class=&p&&{&/span&&span class=&mi&&1&/span&&span class=&p&&};&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&n&&tc2&/span&&span class=&p&&(&/span&&span class=&nb&&j&/span&&span class=&p&&)=&/span&&span class=&n&&tree1&/span&&span class=&p&&.&/span&&span class=&n&&value&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&n&&ratio&/span&&span class=&p&&=&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&nb&&find&/span&&span class=&p&&(&/span&&span class=&n&&tc2&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span&&span class=&p&&))&/span&&span class=&o&&/&/span&&span class=&n&&ntree&/span&&span class=&p&&;&/span&
&span class=&c&&%这里仅有0-1两分类，如有更多分类本句应改写 &/span&
&span class=&k&&if&/span& &span class=&n&&ratio&/span&&span class=&o&&&&/span&&span class=&mf&&0.5&/span&
&span class=&c&&%如判断为上涨的数量更多 &/span&
&span class=&n&&classify2&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)=&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&n&&tp2&/span&&span class=&p&&=&/span&&span class=&mi&&0&/span&&span class=&p&&;&/span&&span class=&n&&fp2&/span&&span class=&p&&=&/span&&span class=&mi&&0&/span&&span class=&p&&;&/span&&span class=&n&&tn2&/span&&span class=&p&&=&/span&&span class=&mi&&0&/span&&span class=&p&&;&/span&&span class=&n&&fn2&/span&&span class=&p&&=&/span&&span class=&mi&&0&/span&&span class=&p&&;&/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&n2&/span&
&span class=&k&&if&/span& &span class=&n&&classify2&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span&
&span class=&k&&if&/span& &span class=&n&&testclass&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span&
&span class=&n&&tp2&/span&&span class=&p&&=&/span&&span class=&n&&tp2&/span&&span class=&o&&+&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&k&&else&/span&
&span class=&n&&fp2&/span&&span class=&p&&=&/span&&span class=&n&&fp2&/span&&span class=&o&&+&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&k&&else&/span&
&span class=&k&&if&/span& &span class=&n&&testclass&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span&
&span class=&n&&fn2&/span&&span class=&p&&=&/span&&span class=&n&&fn2&/span&&span class=&o&&+&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&k&&else&/span&
&span class=&n&&tn2&/span&&span class=&p&&=&/span&&span class=&n&&tn2&/span&&span class=&o&&+&/span&&span class=&mi&&1&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&n&&accu2&/span&&span class=&p&&=(&/span&&span class=&n&&tp2&/span&&span class=&o&&+&/span&&span class=&n&&tn2&/span&&span class=&p&&)&/span&&span class=&o&&/&/span&&span class=&p&&(&/span&&span class=&n&&tp2&/span&&span class=&o&&+&/span&&span class=&n&&tn2&/span&&span class=&o&&+&/span&&span class=&n&&fp2&/span&&span class=&o&&+&/span&&span class=&n&&fn2&/span&&span class=&p&&);&/span&
&span class=&n&&prec2&/span&&span class=&p&&=&/span&&span class=&n&&tp2&/span&&span class=&o&&/&/span&&span class=&p&&(&/span&&span class=&n&&tp2&/span&&span class=&o&&+&/span&&span class=&n&&fp2&/span&&span class=&p&&);&/span&
&span class=&n&&reca2&/span&&span class=&p&&=&/span&&span class=&n&&tp2&/span&&span class=&o&&/&/span&&span class=&p&&(&/span&&span class=&n&&tp2&/span&&span class=&o&&+&/span&&span class=&n&&fn2&/span&&span class=&p&&);&/span&
&span class=&n&&testindex&/span&&span class=&p&&=[&/span&&span class=&n&&accu2&/span&&span class=&p&&,&/span&&span class=&n&&prec2&/span&&span class=&p&&,&/span&&span class=&n&&reca2&/span&&span class=&p&&];&/span&
&/code&&/pre&&/div&&p&&br&&/p&&p&&b&3.策略代码：&/b&&/p&&div class=&highlight&&&pre&&code class=&language-matlab&&&span&&/span&&span class=&p&&[&/span&&span class=&n&&color&/span&&span class=&p&&=&/span&&span class=&n&&black&/span&&span class=&p&&]&/span&&span class=&k&&function&/span& &span class=&n&&randforest&/span&&span class=&p&&(&/span&&span class=&n&&bInit&/span&&span class=&p&&,&/span&&span class=&n&&bDayBegin&/span&&span class=&p&&,&/span&&span class=&n&&cellPar&/span&&span class=&p&&)&/span&
&span class=&c&&%变量声明&/span&
&span class=&n&&traderSetParalMode&/span&&span class=&p&&(&/span&&span class=&n&&false&/span&&span class=&p&&);&/span&
&span class=&k&&global&/span& &span class=&n&&g_idxK&/span&&span class=&p&&;&/span& &span class=&c&&%生成注册日期序列 &/span&
&span class=&k&&global&/span& &span class=&n&&g_idxSignal&/span&&span class=&p&&;&/span& &span class=&c&&%买入标的索引 &/span&
&span class=&k&&global&/span& &span class=&n&&ema12&/span& &span class=&n&&ema26&/span& &span class=&n&&dea&/span& &span class=&n&&macd&/span&&span class=&p&&;&/span&
&span class=&k&&global&/span& &span class=&n&&Tlen&/span&&span class=&p&&;&/span&
&span class=&n&&rforest&/span&&span class=&p&&=&/span&&span class=&n&&cellPar&/span&&span class=&p&&{&/span&&span class=&mi&&1&/span&&span class=&p&&};&/span&
&span class=&k&&if&/span& &span class=&n&&bInit&/span&
&span class=&n&&g_idxK&/span&&span class=&p&&=&/span&&span class=&n&&traderRegKData&/span&&span class=&p&&(&/span&&span class=&s&&'day'&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span&
&span class=&n&&Tlen&/span&&span class=&p&&=&/span&&span class=&nb&&size&/span&&span class=&p&&(&/span&&span class=&n&&g_idxK&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span& &span class=&c&&%获取标的个数 &/span&
&span class=&n&&ema12&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&n&&Tlen&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span&&span class=&n&&ema26&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&n&&Tlen&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span&&span class=&n&&dea&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&n&&Tlen&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span&&span class=&n&&macd&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&n&&Tlen&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span&
&span class=&n&&g_idxSignal&/span&&span class=&p&&=&/span&&span class=&n&&traderRegUserIndi&/span&&span class=&p&&(@&/span&&span class=&n&&RF&/span&&span class=&p&&,{&/span&&span class=&n&&g_idxK&/span&&span class=&p&&,&/span&&span class=&n&&rforest&/span&&span class=&p&&,&/span&&span class=&n&&macd&/span&&span class=&p&&});&/span&
&span class=&k&&else&/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&Tlen&/span&
&span class=&n&&dataDay&/span&&span class=&p&&=&/span&&span class=&n&&traderGetRegKData&/span&&span class=&p&&(&/span&&span class=&n&&g_idxK&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,:),&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&true&/span&&span class=&p&&);&/span& &span class=&c&&%取当日数据&/span&
&span class=&n&&ema12&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&)=&/span&&span class=&mi&&11&/span&&span class=&o&&/&/span&&span class=&mi&&13&/span&&span class=&o&&*&/span&&span class=&n&&ema12&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&)&/span&&span class=&o&&+&/span&&span class=&mi&&2&/span&&span class=&o&&/&/span&&span class=&mi&&13&/span&&span class=&o&&*&/span&&span class=&n&&dataDay&/span&&span class=&p&&(&/span&&span class=&mi&&5&/span&&span class=&p&&,&/span&&span class=&k&&end&/span&&span class=&p&&);&/span&
&span class=&n&&ema26&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&)=&/span&&span class=&mi&&25&/span&&span class=&o&&/&/span&&span class=&mi&&27&/span&&span class=&o&&*&/span&&span class=&n&&ema26&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&)&/span&&span class=&o&&+&/span&&span class=&mi&&2&/span&&span class=&o&&/&/span&&span class=&mi&&27&/span&&span class=&o&&*&/span&&span class=&n&&dataDay&/span&&span class=&p&&(&/span&&span class=&mi&&5&/span&&span class=&p&&,&/span&&span class=&k&&end&/span&&span class=&p&&);&/span&
&span class=&n&&dif&/span&&span class=&p&&=&/span&&span class=&n&&ema12&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&)&/span&&span class=&o&&-&/span&&span class=&n&&ema26&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span&
&span class=&n&&dea&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&)=&/span&&span class=&mi&&8&/span&&span class=&o&&/&/span&&span class=&mi&&10&/span&&span class=&o&&*&/span&&span class=&n&&dea&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&)&/span&&span class=&o&&+&/span&&span class=&mi&&2&/span&&span class=&o&&/&/span&&span class=&mi&&10&/span&&span class=&o&&*&/span&&span class=&n&&dif&/span&&span class=&p&&;&/span&
&span class=&n&&macd&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&)=&/span&&span class=&mi&&2&/span&&span class=&o&&*&/span&&span class=&p&&(&/span&&span class=&n&&dif&/span&&span class=&o&&-&/span&&span class=&n&&dea&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&));&/span&
&span class=&k&&end&/span&
&span class=&p&&[&/span&&span class=&n&&barNum&/span&&span class=&p&&,&/span&&span class=&o&&~&/span&&span class=&p&&]=&/span&&span class=&n&&traderGetCurrentBarV2&/span&&span class=&p&&();&/span&
&span class=&k&&if&/span& &span class=&n&&barNum&/span&&span class=&o&&&&/span&&span class=&mi&&31&/span&
&span class=&k&&return&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&k&&if&/span& &span class=&nb&&mod&/span&&span class=&p&&(&/span&&span class=&n&&barNum&/span&&span class=&o&&-&/span&&span class=&mi&&31&/span&&span class=&p&&,&/span&&span class=&mi&&20&/span&&span class=&p&&)&/span&&span class=&o&&==&/span&&span class=&mi&&0&/span& &span class=&c&&%每20个交易日换仓一次&/span&
&span class=&n&&dSignal&/span&&span class=&p&&=&/span&&span class=&n&&traderGetRegUserIndi&/span&&span class=&p&&(&/span&&span class=&n&&g_idxSignal&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span& &span class=&c&&%获取交易信号&/span&
&span class=&n&&nshare&/span&&span class=&p&&=&/span&&span class=&n&&sum&/span&&span class=&p&&(&/span&&span class=&n&&dSignal&/span&&span class=&p&&);&/span&
&span class=&k&&if&/span& &span class=&n&&nshare&/span&&span class=&o&&==&/span&&span class=&mi&&0&/span& &span class=&c&&%表明没有找到可交易的股票,则维持原仓位不变&/span&
&span class=&k&&return&/span&&span class=&p&&;&/span&
&span class=&k&&end&/span&
&span class=&p&&[&/span&&span class=&n&&map&/span&&span class=&p&&]=&/span&&span class=&n&&traderGetAccountPositionV2&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&Tlen&/span&&span class=&p&&);&/span& &span class=&c&&%获取当前的持仓状况并平仓&/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&Tlen&/span&
&span class=&k&&if&/span& &span class=&n&&map&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)&/span&&span class=&o&&~=&/span&&span class=&mi&&0&/span&
&span class=&n&&traderPositionToV2&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&mi&&0&/span&&span class=&p&&,&/span&&span class=&mi&&0&/span&&span class=&p&&,&/span&&span class=&s&&'market'&/span&&span class=&p&&,&/span&&span class=&s&&'close'&/span&&span class=&p&&);&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&p&&[&/span&&span class=&o&&~&/span&&span class=&p&&,&/span&&span class=&n&&MarketCap&/span&&span class=&p&&]=&/span&&span class=&n&&traderGetAccountInfoV2&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span& &span class=&c&&%获取账户价值 &/span&
&span class=&n&&percash&/span&&span class=&p&&=&/span&&span class=&nb&&floor&/span&&span class=&p&&(&/span&&span class=&n&&MarketCap&/span&&span class=&o&&/&/span&&span class=&n&&nshare&/span&&span class=&p&&);&/span& &span class=&c&&%投资于每只股票的资金 &/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&Tlen&/span&
&span class=&k&&if&/span& &span class=&n&&dSignal&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&)&/span&&span class=&o&&==&/span&&span class=&mi&&1&/span&
&span class=&n&&dataDay&/span&&span class=&p&&=&/span&&span class=&n&&traderGetRegKData&/span&&span class=&p&&(&/span&&span class=&n&&g_idxK&/span&&span class=&p&&(&/span&&span class=&nb&&i&/span&&span class=&p&&,:),&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&true&/span&&span class=&p&&);&/span& &span class=&c&&% 取当日数据 &/span&
&span class=&n&&sharenum&/span&&span class=&p&&=&/span&&span class=&nb&&floor&/span&&span class=&p&&(&/span&&span class=&n&&percash&/span&&span class=&o&&/&/span&&span class=&p&&(&/span&&span class=&n&&dataDay&/span&&span class=&p&&(&/span&&span class=&mi&&5&/span&&span class=&p&&,&/span&&span class=&k&&end&/span&&span class=&p&&)&/span&&span class=&o&&*&/span&&span class=&mi&&100&/span&&span class=&p&&))&/span&&span class=&o&&*&/span&&span class=&mi&&100&/span&&span class=&p&&;&/span&
&span class=&n&&traderBuyV2&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&nb&&i&/span&&span class=&p&&,&/span&&span class=&n&&sharenum&/span&&span class=&p&&,&/span&&span class=&mi&&0&/span&&span class=&p&&,&/span&&span class=&s&&'market'&/span&&span class=&p&&,&/span&&span class=&s&&'long'&/span&&span class=&p&&);&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&k&&end&/span&
&span class=&c&&%---------------------&/span&
&span class=&k&&function&/span&&span class=&w&& &/span&value0&span class=&p&&=&/span&&span class=&nf&&RF&/span&&span class=&p&&(&/span&cellPar,bpPFCell&span class=&p&&)&/span&&span class=&w&&&/span&
&span class=&n&&idxK&/span&&span class=&p&&=&/span&&span class=&n&&cellPar&/span&&span class=&p&&{&/span&&span class=&mi&&1&/span&&span class=&p&&};&/span&
&span class=&n&&rforest&/span&&span class=&p&&=&/span&&span class=&n&&cellPar&/span&&span class=&p&&{&/span&&span class=&mi&&2&/span&&span class=&p&&};&/span&
&span class=&n&&macd&/span&&span class=&p&&=&/span&&span class=&n&&cellPar&/span&&span class=&p&&{&/span&&span class=&mi&&3&/span&&span class=&p&&};&/span&
&span class=&n&&ntree&/span&&span class=&p&&=&/span&&span class=&nb&&length&/span&&span class=&p&&(&/span&&span class=&n&&rforest&/span&&span class=&p&&);&/span&
&span class=&n&&targetNum&/span&&span class=&p&&=&/span&&span class=&nb&&size&/span&&span class=&p&&(&/span&&span class=&n&&idxK&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&);&/span&
&span class=&n&&value0&/span&&span class=&p&&=&/span&&span class=&nb&&zeros&/span&&span class=&p&&(&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&n&&targetNum&/span&&span class=&p&&);&/span&
&span class=&p&&[&/span&&span class=&o&&~&/span&&span class=&p&&,&/span&&span class=&o&&~&/span&&span class=&p&&,&/span&&span class=&o&&~&/span&&span class=&p&&,&/span&&span class=&o&&~&/span&&span class=&p&&,&/span&&span class=&n&&hs300&/span&&span class=&p&&]=&/span&&span class=&n&&traderGetKData&/span&&span class=&p&&(&/span&&span class=&s&&'sse'&/span&&span class=&p&&,&/span&&span class=&s&&'000300'&/span&&span class=&p&&,&/span&&span class=&s&&'day'&/span&&span class=&p&&,&/span&&span class=&mi&&1&/span&&span class=&p&&,&/span&&span class=&mi&&0&/span&&span class=&o&&-&/span&&span class=&mi&&31&/span&&span class=&p&&,&/span&&span class=&mi&&0&/span&&span class=&p&&,&/span&&span class=&n&&true&/span&&span class=&p&&,&/span&&span class=&s&&'FWard'&/span&&span class=&p&&);&/span& &span class=&c&&%获取沪深300指数&/span&
&span class=&k&&for&/span& &span class=&nb&&i&/span&&span class=&p&&=&/span&&span class=&mi&&1&/span&&span class=&p&&:&/span&&span class=&n&&targetNum&/span&
&span class=&n&&regKMatrix&/span&&span class=&p&&=&/span&&span class=&n&&traderGetRegKData&/span&&span class=&p&&(&/span&&span class=&n&&idxK&/span&&