box cox转换
I recently started (re-)reading Forecasting: Principles and Practices to get a quick refresher on (you’ll never guess) forecasting!
我最近开始(重新)阅读《 预测:原理和实践》 ,以快速复习(您永远不会猜到)预测!
The textbook is short, easy for beginners, and a wonderful appendage to the monstrous mess of books, YouTube videos, and MOOCs that aspiring data scientists simply “must” learn.
该教科书简短易懂,适合初学者使用,它是书籍,YouTube视频和MOOC的绝妙补充,那些有抱负的数据科学家只需“必须”学习。
More importantly, the textbook is free.
更重要的是,这本教科书是免费的。
The authors of Forecasting devote one sub-chapter to transforming data (Section 3.2: “Transformations and Adjustments”), where they go over four types of transformations.
预测的作者将一个子章节专门用于转换数据(第3.2节:“转换和调整”),其中介绍了四种转换类型。
One of these transformations (and the first I was introduced to in undergrad) is the Box-Cox Transformation.
Box-Cox转换是其中的一种转换(也是我在本科生中第一次介绍的)。
Why Would We Want to Transform Our Data?
为什么我们要转换数据?
The Box-Cox transformation transforms our data so that it closely resembles a normal distribution.
Box-Cox转换会转换我们的数据,使其与正态分布非常相似。
In many statistical techniques, we assume that the errors are normally distributed. This assumption allows us to construct confidence intervals and conduct hypothesis tests. By transforming your target variable, we can (hopefully) normalize our errors (if they are not already normal).
在许多统计技术中,我们假设误差是正态分布的。 这个假设使我们能够构建置信区间并进行假设检验。 通过转换目标变量,我们可以(希望)对错误进行归一化(如果它们还不是正常的话)。
Additionally, transforming our variables can improve the predictive power of our models because transformations can cut a