automate my entire analysis?
ReservePro Enterprise Edition (RPEE) divides your regular analysis process in three different sections or thought processes:
| · What data is to be analyzed; for example: | |
| all commercial lines, by reserving segment, for each company | |
| · How each data set should be analyzed; for example: | |
| which exhibits and methods are appropriate how development factors and other selections are made how tail factors are estimated how different method calculations are weighted and used to arrive at selected ultimates |
|
| · What processes are involved with each data set; for example: | |
| compressing monthly data to quarterly triangles for analysis selecting development and tail factors selecting ultimates printing reports saving the final analysis |
|
RPEE uses three system components to handle these different parts of the process:
| Queries | tell RPEE which data sets to retrieve from the database for analysis, and in what order | ||
| Example: | |||
| · | all loss, expense, count, and related data triangles for all commercial lines for companies A, B, C and D as of 12/31/02 | ||
| Templates | define the calculation rules for how basic analysis functions should be performed | ||
| Examples: | |||
| · | Use a Volume Weighted Average when making selections for the Incurred Loss Development and Paid Loss Development methods. | ||
| · | Select the related tail factors based on the best fit of three different curves fit through the selected factors. | ||
| · | Append industry development factors to the exhibits for comparison purposes. | ||
| · | When selecting ultimate losses, use a weighted average of Paid and Incurred Loss Development methods and a Bornhuetter-Fergusson method. | ||
| Scripts | list all the steps to be performed with each data set | ||
| Examples: | |||
| · | Compress monthly data to quarterly triangles for analysis | ||
| · | Make development factor and tail factor selections according to Template guidelines | ||
| · | Make ultimate selections according to Template guidelines | ||
| · | Print a predefined list of exhibits and reports | ||
| · | Save all the data and selections | ||
| · | Export key selections to a separate table that retains summary information from across all analyses | ||
RPEE combines queries, templates, and scripts into "projects" to automate large portions of your analysis for you. Following is an example of how this automation process might typically work:
| Step 1: | RPEE evaluates the query, and determines which data set to load first from the database. It loads all data for Workers Comp for Company A, including arrays for losses, expenses, premiums, claim counts, exposures, and subrogation. |
| Step 2: | The system makes initial loss development factor selections, selecting the Volume Weighted Average factors for the Incurred Loss Development and Paid Loss Development methods. Tail factors are selected by fitting the Inverse Power Curve to the selected factors. |
| Step 3: | Ultimate losses are selected from Paid and Incurred Loss Development methods and a B-F method, based on a predefined weighting among the methods by accident period. |
| Step 4: | The system prints arrays, methods, and summary reports to document the analysis of this segment of the business. |
| Step 5: | Finally, the script instructs the system to save the data and selections in the database. |
| Step 6: | If this is the last step in the script, RPEE now starts over, retrieving the next data set from the database (for example, General Liability for Company A) and beginning the process again. It will continue this processing until all data sets in the query have been evaluated. |
Through this process, the system could go through a "first cut" of a significant portion of your regular analysis. You can then focus your attention on the areas that require your expertise, while reviewing the other portions for reasonableness.
Also note that, while the entire process can be automated to run as described above, you can also set "breaks" in the scripts, temporarily halting the system at specific points in the process. With each data set, you could stop, quickly review the automated selections, make minor adjustments if necessary, and then let the process continue. This way, you can work through much of your analysis very quickly, while still involving your professional judgment along the way.
Please feel free to contact us if you'd like to see how ReservePro Enterprise easily automates much of your routine analysis work for you. Finally, you can spend your time doing the things that really matter to you and your department!