| |
:: Course SummaryThis half-day workshop offers an overview of tree-based modeling techniques, starting with the interpretation of results, strengths and weaknesses and also discusses their practical realization.:: Learning ObjectivesUpon completion of this training course, participants will have learned:
- How to build and interpret classification and regression trees
- How to measure the performance and fit of a regression tree, and how to improve it
- The advantages as well as the downsides of using this technique
:: Target AudienceThis applied training session in statistics is aimed at all scientific staff who collect data and who must make decisions based on them. The regression techniques covered in this session will be particularly useful for people who are interested in exploring a new way of relating/predicting a variable to/from a set of explanatory variables.:: PrerequisiteParticipants should know the essential tools in statistics - descriptive statistics, both numerical (mean, standard deviation, standard error, etc.) and graphical (histogram, box-plot, scatter plot, etc.), hypothesis testing and confidence intervals
Either by having attended the training session Fundamental Tools in Statistics or by possessing a similar background.
Working knowledge of ordinary multiple regression techniques is desirable but not mandatory. | | |
:: Topics Covered
- A Different Principle
- C&RT methods as an alternative to linear regression
- A few examples
- Interpretation and use of tree-based models
Models for a continuous response
Models for a discrete response
- Basic Model-Building: Tree Growing
- Criteria and algorithms for selecting optimal split
- Constraints on node and leave size
- Modifying control parameters
- Model Improvement: Pruning the tree
- Reasons for pruning
- Pruning methods
- Model selection
Crossvalidation and alternative techniques
Selecting a number of nodes
- Pruning output
- Detailed Output
- Understanding and interpreting software output for tree models
Tables and summary statistics
- Final Model Performance and Stability
- C&RT methods versus classical regression performance
- Sample size considerations
- Recent tools to improve the model stability: bagging and boosting
- Advanced Methods
- Using trees for prediction purposes
- Combining tree-based techniques with classical regression tools
- An Overview of Other Tree-Based Methods
:: Course ContentThis half-day session covers classification and regression tree methods, another way to related a variable to a set of explanatory variables.
The training covers the objectives of the tree-based regression analysis , the principle of modeling, parameter estimation, as well as interpretation of model coefficients, goodness-of-fit and validation measures, prediction in regression, problems commonly encountered in regression, ways to detect them and to solve them.
| |
