正在学习
I Foundations
I Foundations
1 Introduction
1.1 Outline of the Book
1.2 Technical Notes
2 Loss Functions
2.1 Construction and Specification of the Loss Function
2.2 Specific Loss Functions
2.3 Multivariate Loss Functions
2.4 Scoring Rules for Distribution Forecasts
2.5 Examples of Applications of Forecasts in Macroeconomics and Finance
2.6 Conclusion
3 The Parametric Forecasting Problem
3.1 Optimal Point Forecasts
3.2 Classical Approach
3.3 Bayesian Approach
3.4 Relating the Bayesian and Classical Methods
3.5 Empirical Example: Asset Allocation with Parameter Uncertainty
3.6 Conclusion
4 Classical Estimation of Forecasting Models
4.1 Loss-Based Estimators
4.2 Plug-In Estimators
4.3 Parametric versus Nonparametric Estimation Approaches
4.4 Conclusion
5 Bayesian Forecasting Methods
5.1 Bayes Risk
5.2 Ridge and Shrinkage Estimators
5.3 Computational Methods
5.4 Economic Applications of Bayesian Forecasting Methods
5.5 Conclusion
6 Model Selection
6.1 Trade-Offs in Model Selection
6.2 Sequential Hypothesis Testing
6.3 Information Criteria
6.4 Cross Validation
6.5 Lasso Model Selection
6.6 Hard versus Soft Thresholds: Bagging
练习题
Which of the following is NOT a key aspect of constructing and specifying a loss function?
A. Defining the decision space
B. Specifying the parameter space
C. Determining the forecast horizon
D. Choosing the optimal forecast method
Which loss function is commonly used for regression problems and is defined as ?
A. Absolute error loss
B. Quadratic loss
C. Zero-one loss
D. Log loss
Which of the following are characteristics of multivariate loss functions? (Select all that apply)
A. They handle multiple dependent variables simultaneously
B. They are only used for classification problems
C. They can incorporate correlations between variables
D. They are defined for a single output variable
E. They are used to evaluate the performance of multivariate forecasts
Which of the following are scoring rules for distribution forecasts? (Select all that apply)
A. Logarithmic score
B. Quadratic score
C. Zero-one score
D. Spherical score
E. Absolute error score
Forecasts in macroeconomics and finance are primarily used for short-term predictions only.
The application of forecasts in finance often involves risk management and portfolio optimization.
The optimal point forecast under quadratic loss is the ___ of the conditional distribution of given .
In the classical approach to forecasting, the parameters are estimated using ___ methods such as ordinary least squares.
Explain the key difference between the Bayesian and classical approaches to forecasting.
How do the Bayesian and classical methods relate to each other in the context of forecasting?
In the empirical example of asset allocation with parameter uncertainty, which approach is used to account for the uncertainty in the parameter estimates?
A. Classical approach only
B. Bayesian approach only
C. Both classical and Bayesian approaches
D. Neither approach accounts for parameter uncertainty
Which type of estimator is derived by minimizing the expected loss function with respect to the parameters?
A. Plug-in estimator
B. Maximum likelihood estimator
C. Loss-based estimator
D. Moment estimator
Plug-in estimators are always consistent estimators of the true parameters.
Parametric estimation approaches assume a specific functional form for the ___, while nonparametric approaches do not.
Define Bayes risk and explain its significance in Bayesian forecasting.
Which of the following are examples of ridge and shrinkage estimators? (Select all that apply)
A. Lasso estimator
B. Ridge regression estimator
C. Ordinary least squares estimator
D. Elastic net estimator
E. Maximum likelihood estimator
登录后解锁笔记、知识点解析、AI 问答
立即登录