Exam PA – Predictive Analytics
5 个单元 · 22 个章节
免费 Exam PA – Predictive Analytics 学习笔记，适合 SOA (Society of Actuaries) 学生。每个章节都覆盖一个重点主题，并附例题与可延伸到 thinka app 的练习提示。
Predictive Analytics Problem Definition
Descriptive, predictive, and prescriptive analytics
Characteristics of predictive modeling problems
Bias, variance, model complexity, and the bias-variance trade-off
Translating a vague question into an analyzable one
Defining the problem: data, technology, business impact, and implementation
Identifying additional information and next steps
Data Exploration and Visualization
Structured and unstructured data types
Types of variables and predictive modeling terminology
Data design: time frame, sampling, and granularity
Key principles of constructing graphs
Univariate data exploration
Bivariate data exploration
Data Transformations and Unsupervised Learning Techniques
Creating features from existing data
Principal components analysis for data transformation
K-means and hierarchical clustering for data transformation
Generalized Linear Models
Selecting and validating a GLM for a business problem
Offsets and weights
Interpreting model coefficients and interaction terms
Hyperparameters for regularized regression
Tree-Based Models
Constructing, pruning, and validating regression and classification trees
Bagging and random forests
Boosting
看完笔记了？现在就用 AI 题目测一测自己。
马上练这一题