Exam PA – Predictive Analytics
5個單元 · 22個課題
免費的Exam PA – Predictive Analytics學習筆記，專為SOA (Society of Actuaries)學生準備。下列每個章節都涵蓋一個重點主題，附有例題與練習提示，可在 thinka 應用程式中延伸練習。
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
看完筆記了？用AI練習題測試自己
立即練習此課題