漫步量化

OverviewBarra risk factors are particular factors used to implement the Barra risk factor analysis, that consists in a multi-factor model developed by Barra inc. to measure the overall risk to which a financial asset is exposed.From investopedia.com, t..

DecileFrom wikipedia, in descriptive statistics, a decile is any of the nine values that divide the sorted data into ten equal parts, so that each part represents 1/10 of the sample or population.A decile is one possible form of a quantile, others incl..

Missing dataMissing data can grocely be classified into three types:MCAR(Missing Completely At Random), which means that there is nothing systematic about why some date is missing. That is, there is no relationship between the fact that data is missin..

异常值检测Detect OutliersIn statistics, outliers are data points that don’t belong to a certain population. It is an abnormal observation that lies far away from other values. An outlier is an observation that diverges from othervise well-structured data.Th..

Feature scalingFeature scaling is a method used to normalize the range of independent variables or features of data.In some machine learning algorithms, objective functions will not work without normalization.Gradient descent converges much faster wit..

Risk ExposureRisk Exposure is the measure of potential future loss resulting from a specific activity or event.Finiancial ExposureIn finance, exposure refers to the amount of money that an investor has invested in a particular asset.It represents t..

Risk ExposureRisk Exposure is the measure of potential future loss resulting from a specific activity or event.Finiancial ExposureIn finance, exposure refers to the amount of money that an investor has invested in a particular asset.It represents t..

Investment Tools HistoryTimeStatesInvestment Tools1950之前1. There was no concept of systematic, or market-related, return.2. Return was a rise in the value of an asset and risk was a drop in the value of an asset.intuition;insightful financi..

www.mscibarra.comMSCI (Morgan Stanley Capital International and MSCI Barra), is an American finance company headquartered in New York City and serving as global provider of equity, fixed income, hedge fund stock market indexex, and multi-asset portfolio..

CAPM与APTCAPMCAPMCAPM（Capital  Asset  Pricing  ModelCapital \;Asset \;Pricing \; ModelCapitalAssetPricingModel，资本资产定价模型）；APTAPTAPT（Arbitrage  Pricing  TheoryArbitrage \; Pricing \; TheoryArbitrage...

关键字关键字关键字阻力支撑相对强度RSRS参数敏感性测试Resistance Support Relative StrengthRSRS右偏标准分指标RSRS价格优化右偏标准分策略支撑位与阻力位阻力位与支撑位的运用思路大多在突破策略与反转策略中。常见确定方式有：通道线（布林带的上下轨）；一段时间内的前高与前低；区间震荡线；主...

多因子模型的构建方式通常被分为：打分法、回归法打分法打分法是指选用若干能够对股票收益产生预测作用的因子，之后根据股票的每个因子值在截面上的相对位置给出股票在该因子上的得分，然后按照一定的权重将每个股票的各个因子得分相加从而得到该股票的最终得分并按照该得分对股票进行排序，筛选，构造投资组合。打分法的关键要点有：因子的权重即便因子很好，如果权重不合适，整体结果可能也会很糟糕。常见的权...

接触多因子策略，总会看到IC值、IR值，作为某种度量指标。IC值的定义IC是***Information Coefficient***的缩写，称为***信息系数***。IC代表的是预测值和实现值之间的相关性，通常用于评价预测能力（即选股能力）。IC∈[−1,  1]绝对值越大，表示预测能力越好IC \in [-1,\; 1] \\ 绝对值越大，表示预测能...

(一)Barra 模型初探：A股股市场风格解析2018.1.6Barra风险-收益归因模型发展历史多因子模型的基础理论认为：股票的收益是由一些共同的因子来驱动的，不能被这些因子解释的部分被称为股票的“特质收益率”，而每支股票的特质收益率之间是互不相关的。Barr Rosenberg(1974)提出采用多因子风险模型来对投资组合的风险和收益进行分析。并成立Barra，后提出Barra ...

对于量化研究而言，拿到一份基础数据，首先需要对数据做预处理，以便于更好的探究数据规律，基于不同目标有不同处理环节，其中去极值、标准化、中性化这三点经常被放在一起讨论。整理网络资料，理解如下。去极值数据（单个因子的时间序列数据）中存在异常值，可能会导致拉大标准差、拉偏统计规律…常规处理理念就是确定此数据指标（比如某个因子）的上下限阈值，然后超过（或低于）此限度的数据均设置为阈值，以提高数据...

(一)因子测试框架2017.4.10根据现代金融理论的定义，投资组合获取的收益分为两部分：来自市场的收益(Beta) + 超出市场的收益(Alpha)。获取Alpha的理论有：资本资产定价模型(CAPM)、Fama-French三因素模型、基于套利定价理论(APT)的多因子模型。多因子模型从构建目标角度分为：Alpha模型风险模型。以Barra为代表的风险模型更多的用于投资组...

做量化，经常听到多因子模型，豆瓣直接搜索“多因子模型”找不到相关的理论书籍，汇总网络资料，大概搞清楚什么是多因子模型，以及多因子模型的理论基础在哪里。什么是多因子定价模型？APT（套利定价理论）、Fama-French三因子模型之间的关系是怎样的？多因子资产定价模型知识对定价核(Pricing Kernel)的线性逼近，ICAPM，APT和Fama-French是三个最流行的分支。多...