Business Context
A global logistics company has fitted a complex dependency model between 'Shipping Delay' and 'Fuel Cost' across varying routes. The Data Engineering team needs to export the complete model definition and all optimizer settings to document the model version 2.0 in their governance system.
Data Preparation
Generating a larger dataset representing 10,000 shipping routes.
Copied!
| 1 | |
| 2 | DATA mycas.logistics_data; |
| 3 | DO i = 1 to 10000; |
| 4 | delay = rand('Exponential'); |
| 5 | cost = delay * 1.5 + rand('Uniform'); |
| 6 | OUTPUT; |
| 7 | END; |
| 8 | |
| 9 | RUN; |
| 10 |
Étapes de réalisation
1
Fit a Clayton Copula (suitable for tail dependence) and save to 'logisticsModel_v2'.
Copied!
| 1 | |
| 2 | PROC CAS; |
| 3 | copula.copulaFit TABLE={name='logistics_data'} copula='CLAYTON' var={'delay', 'cost'} store={name='logisticsModel_v2', replace=true}; |
| 4 | |
| 5 | RUN; |
| 6 |
2
Retrieve ALL information from the store but use the 'display' parameter to only print the 'ModelDefinition' to the log to reduce verbosity, while creating all default outputs.
Copied!
| 1 | |
| 2 | PROC CAS; |
| 3 | copula.copulaViewStore / instore={name='logisticsModel_v2'} viewOptions={all=true} display={names={'ModelDefinition'}}; |
| 4 | |
| 5 | RUN; |
| 6 |
Expected Result
The action accesses 'logisticsModel_v2' and prepares all model data internally (due to viewOptions={all=true}). However, in the user results, only the 'ModelDefinition' table is visually displayed, confirming the Clayton copula type and parameter constraints.