随笔分类 -  ISLR

ISLR第8章The Basics of Decision Trees
摘要：The Basics of Decision Trees In this chapter, we describe tree-based methods for regression and classification. These involve stratifying or segmentin 阅读全文
posted @ 2018-04-19 16:04 weiququ 阅读(512) 评论(0) 推荐(0)
ISLR第10章 Unsupervised Learning
摘要：Unsupervised learning, a set of statistical tools intended for the setting in which we have only a set of featuresX1,X2, . . . , Xp measured on n obse 阅读全文
posted @ 2018-03-25 21:56 weiququ 阅读(405) 评论(0) 推荐(0)
ISLR第9章SVM
摘要：In this chapter, we discuss the support vector machine (SVM), an approachfor classification that was developed in the computer science community inthe 阅读全文
posted @ 2018-03-15 14:52 weiququ 阅读(482) 评论(0) 推荐(0)
ISLR第六章Linear Model Selection and Regularization
摘要：本章主要介绍几种可替代普通最小二乘拟合的其他一些方法。 Why might we want to use another fitting procedure instead of least squares? better prediction accuracy（预测精度） and better m 阅读全文
posted @ 2018-03-06 18:35 weiququ 阅读(1087) 评论(0) 推荐(0)
ISLR第五章Resampling Methods（重抽样方法）
摘要：Resampling methods are an indispensable tool in modern statistics. In this chapter, we discuss two of the most commonly used resampling methods, cross 阅读全文
posted @ 2018-03-06 16:12 weiququ 阅读(2899) 评论(0) 推荐(0)
ISLR第二章
摘要：1 Statistical Learning 1.1 What Is Statistical Learning? More generally, suppose that we observe a quantitative response Y and p different predictors, 阅读全文
posted @ 2018-03-06 10:37 weiququ 阅读(567) 评论(0) 推荐(0)
ISLR第四章分类
摘要：1. 为什么线性回归不可用？ 通常，两个以上定性响应变量不能用线性回归建立模型 线性模型不能保证预测值在0，1之间 2. logistic regressiom 2.1 logistic model logistic function: odds: 取值范围为0 到∞。其值接近于0 表示违约概率非常 阅读全文
posted @ 2018-02-09 23:08 weiququ 阅读(539) 评论(0) 推荐(0)
ISLR第3章线性回归
摘要：方程式： 1.1 估计参数 代表第i 个残差第i 个观测到的响应值和第i 个用线性模型预测出的响应值之间的差距 残差平方和（residual sum of squares ,RSS）： 等价于： 最小二乘法选择β0和β1来使RSS达到最小。通过微积分运算，使RSS最小的参数估计值为： 1.2评估系数 阅读全文
posted @ 2018-01-19 18:01 weiququ 阅读(958) 评论(0) 推荐(0)