This section provides information and the functional flow of the Oracle Insurance Loss Modeller Application.
Topics:
· Logging in to the OILM Application
To log in to the OILM Application, perform the following steps:
1. Access the OILM Application by using the login credentials (User ID and Password) provided and select the preferred language to navigate. The built-in security system ensures that you are only permitted to access the window and actions based on the authorization.
2. After logging in to the OFSAA Home screen, the landing page is displayed.
Figure 1: The OILM Landing Page
3. Use the information provided in the following table to set the application preferences.
Field |
Description |
|---|---|
User Menu |
The following options are available in this drop-down: · Preferences · About · Change Password · Logout. |
Application |
Click this icon to view all the applications installed in your environment. |
Language |
This menu displays the language you selected in the OFSAA Login Window. The language options displayed in the Language Menu are based on the language packs installed in your OFSAA instance. Using this menu, you can change the language at any point in time. |
Administration |
Click this icon to navigate to the Administration window. The Administration window displays modules such as: · Translation Tools · Object Administration · Utilities. |
Last Failed Login Date & Time |
Click this icon to view the details of the last login and last failed login. |
Object Administration |
Object Administration is an integral part of the infrastructure and facilitates system administrators to define the security framework. See the OFS Advanced Analytics Infrastructure User Guide for details. |
Common Object Maintenance |
Common Object Maintenance is an integral part of the infrastructure system and facilitates system administrators to define the security framework with the capacity to restrict access to the data and metadata in the warehouse, based on a flexible, fine-grained access control mechanism. See the OFS Advanced Analytics Infrastructure User Guide for details. |
4. Select Oracle Insurance Loss Modeller on the OFSAA landing page.
5. Select Oracle Insurance Loss Modeller in the Left-Hand Side (LHS) pane.
The following diagram depicts the functional flow of the Oracle Insurance Loss Modeller Application:
Figure 2: The Functional Flow of the Oracle Insurance
Loss Modeller Application
The Source Data flows from the Policy Admin and Claim system into the Dashboards
The Policy Admin is a system that has records of all policies written by insurance companies. It performs and stores all the key elements for rating, quote generation, binding, issuing, reinsurance, endorsement, renewals, and so on.
The Claim System is a system that has records of all the claims and related details reported to an insurance company. It stores all the key elements of the claims such as, claim amount, lines of business, coverage details, reported claims, approved claims, declined claims, and so on.
The Operations UI contains the status of the Data Upload, displays the As of Date of Data, Reconciliation of the uploaded data, and maintains the logs of activities. For more information, see the Oracle Financial Services Analytical Applications Infrastructure Administration and Configuration Guide
When configuring the OILM Configuration Tables, the client field is mapped against the fields mentioned in the Dimension Tab. The Flag(Y) Column denotes all the fields the user would like to have displayed on the Dashboard Page as a filter. Additionally, two filters are fixed for every user with a different color in the first row. On the Dashboard Page, all KPIs about the Insurance industry are displayed in the form of graphs. The user has the option to refine them for the Business Unit (Towers), LOB, Product, Sub Product, and at the Coverage Level. For now, the application displays 4 different charts on the screen.
The first graph contains a double Y-axis, on which one Y-axis contains the amount and the other Y-axis contains the percentage. The line graph in this graph represents the loss reports across different timelines. The bar represents the Premium and Claim across the same timeline.
Figure 3: The Dashboard Page
Projects is a library that is used to store all the monthly or quarterly, or annual projection exercises in the Loss Modeller. For the first time, the default project will be stored in the Projects folder. This folder can be renamed and you can work on the default project after uploading the data. From the next term onwards the user needs to Roll-Forward the existing project to create the next term project. For more information on how to use the Methods feature in the application, see the Projects Summary Section.
Figure 4: The Projects Summary Page
The Triangles Page displays the default Cumulative Triangle and it displays the Premium, Expenses, and Losses, on the Dashboard Page. For more information on how to use the Methods feature in the application, see the Triangles Section.
Figure 5: The Triangles Page
The following section details the various triangles available in the application
The Cumulative Triangle is derived from the Incremental Triangle and is a standard way of displaying the subsequent development of Losses and Premium from their Start Dates, Accident, and UW Years in this case. Developments such as Yearly, Monthly, Quarterly, and Half-Yearly are covered in the application.
The Cumulative Triangle is not applicable to the Outstanding, Partly Paid (Amount and Number) Triangles. The Yearly, Monthly, Quarterly, and Half-Yearly developments are applicable to all Triangles.
To derive the Cumulative Triangle from the Incremental Triangle, each value in the Incremental Triangle adds the previous number, the left-hand side number, from the Cumulative Triangle. The number in the second position is the sum of the number in the first position in the Cumulative Triangle, and the number in the second position in the Incremental Triangle in each row.
Figure 5: The Cumulative Triangle
Depending on the development period for the Triangle, the Triangle page will display the following developments:
· Monthly – Monthly Triangle by default displays 1 Year (12 Developments).
· Quarterly – Quarterly Triangles displays 2 years (8 Developments).
· Half yearly – Half-Yearly Triangles displays 5 Years (10 Developments).
· Yearly – Yearly Triangles displays 10 Years (10 Developments).
If the Attritional Paid Loss Triangle is created for a particular line of business, then the column selected from the database is Loss Type with the selected value as Attritional Loss. For the Paid Loss Triangle, the column to be considered is Paid Loss in the Claim System. For the Outstanding Loss Triangle the Outstanding Loss Column is used. For the Gross Premium Triangle, the Premium Column from Policy Admin is used. These columns are in addition to other columns such as Lines of Business, Business Unit, Coverage, and so on based on the selection criteria in the dashboard filters. To create a Triangle, the logic is to first summarize the data based on the filter selection such as Business Unit, Lines of Business, Product, Sub Product, Coverage, Region, Currencies, Loss Type, and so on based on the selected dimension.
By default, multiple varieties of Triangles are pre-configured within the application and these Triangles can be copied or modified but cannot be deleted.
The Triangle can be Draft, Submitted, Approved, or Rejected based on the status of the work progress.
The user that has Admin rights has the option to Approve or Reject a Triangle. These options are only available when the status of a Triangle is Submitted.
Figure 6: The Additional Loss Triangle
The Incremental Triangle represents the losses and premium for the given Accident or UW year or Reporting Cohort respectively for a particular point of time when these triangles were created directly from the data. Developments such as Yearly, Monthly, Quarterly, and Half-Yearly are covered in the application. In case it is a loss Triangle such as Paid or Outstanding and Claim-related expense, the source table will be the Claim system and the Triangle basis (left vertical axis) in the Triangle will be Accident ore, UW, or Reporting Year. The default setting can be modified to another basis. The default setting can be modified to another basis.
Figure 6: The Incremental Triangle
The Attritional Paid Loss Triangle, Paid Loss Triangle, Outstanding Loss Triangle, and Gross Premium Triangle work similarly to the Cumulative triangle in this tab.
The Development Factor Triangle is calculated by using the Cumulative Triangle. Irrespective of a selected Triangle, the Development Factor, and Age to Age Factor are always derived from the Cumulative Triangle.
Figure 27: The Age to Age Factor Triangle
Figure 28: The Average Age to Age Factor Triangle
The following is the formula used for calculating the different averages:
· Simple Average – Latest 5: It calculates the simple average of the latest 5 years that are available in the Development Factor Triangle. Note that if the data is available till the year 2020, then the Development Factor will contain factors only till the year 2019. Based on this understanding the latest year, in this case, will be 2019 and not 2020. The simple 5-year average is calculated by using the years 2015, 2016, 2017, 2018, and 2019 years.
· Simple Average – Latest 3: It calculates the simple average of the latest 3 years available in the Development Factor Triangle. For example, the years 2017, 2018, and 2019.
· Simple Average – Latest 2: It calculates the simple average of the latest 2 years available in the Development Factor Triangle. For example, the years 2018 and 2019.
· Volume weighted average – Latest 5: It calculates the Volume Weighted Average of the latest 5 years that are available in the Development Factor Triangle. It calculates the Weighted Average by using the Development Factors from the Development Triangle and the corresponding numbers from the Cumulative Triangle.
· Volume weighted average – Latest 4: It calculates the Volume Weighted average of the latest 4 years that are available in the Development Factor Triangle. It calculates the Weighted Average by using the Development Factors from the Development Triangle and the corresponding numbers from the Cumulative Triangle.
· Volume weighted average – Latest 3: It calculates the Volume Weighted Average of the latest 3 years available in the Development Factor Triangle. It calculates the Weighted Average by using the Development Factors from the Development Triangle and the corresponding numbers from the Cumulative Triangle.
· Volume weighted average – Latest 2: It calculates the Volume Weighted Average of the latest 2 years that are available in the Development Factor Triangle. It calculates the Weighted Average by using the Development Factors from the Development Triangle and the corresponding numbers from the Cumulative Triangle.
· All-year average: It calculates the Simple Average by using all the available years in the Development Triangle.
· Geometric Average – Latest 3: It calculates the Geometric Average of the latest 3 years available in the Development Factors Triangle.
· Selected: This section displays all the selected averages. This section is editable and can be modified after selecting an average. Users can select any of the above-average by clicking the radio button based on their observation or requirement.
· Cumulative Development Factor: This section calculates the Cumulative Development Factor (CDF). The calculation uses a user-selected row as a base.
· Ratio to Ultimate Factor: The Ratio to Ultimate Factor is calculated based on the Cumulative Development Factor.
The Projection Factor is calculated to get the Development Patterns for future developments (such as future years, half years, quarters, or months). The data is derived from the Development Factor or Age to Age Factor and is displayed in a Triangle format.
Figure 8: The Projection Factor
The Projection Triangle has two sections; the first section is the Cumulative Triangle, which is already generated under the Cumulative Triangle Tab, and the second section is the projection numbers. Here, the Development Factors from the Projection Factor are multiplied by the latest cumulative numbers, brought in here from the Cumulative Triangle, to produce projected numbers.
Figure 8: The Projection Triangle
This section provides detailed information on the Methods that are available in the application.
The Chain Ladder Method is used to forecast the reserve that must be established for a particular year to cover future losses. The exercise uses projected losses from the triangulation method. The Chain Ladder Method requires the Cumulative Paid Triangle and Cumulative Reported Losses Triangle as a prerequisite, however, the application enables the modification of the basis of these calculations and different Triangles can be selected. For more information on how to use the Methods feature in the application, see the Methods Section.
The customized approach is a continuation of the Chain Ladder Method. The first table is the same as the Chain Ladder Method. Two different sets of Adjustments can be added in the Adjustment 1 and Adjustment 2 Columns for the final output.
Both the columns, Adjustment 1 and Adjustment 2, are editable at the cell level, and adding a comment for a modified cell is mandatory.
The calculation of IBNR, IBNR based on Dev. Method with, and Reserve, Total based on Using Dev. Method displays both Adjustment 1 and Adjustment 2 as an Input Variable for customization and adjustments.
The adjustment columns enable the upload of Patterns and other calculations performed within the application as well as external sources such as Excel, CSV, etc as these calculations or Patterns are available for all the Accident Years. Failing this, the system gives an error message informing us that the input is not in the correct format.
The Expected Claim Method is used to forecast the amounts. This is done by projecting historical experiences into the future.
The Expected Claim Method requires the Cumulative Paid Triangle, Cumulative Reported Losses Triangle along with Cumulative Earned Premium Triangle as a prerequisite.
On clicking the Expected Claim Tab on the Methods Page, three tables will be displayed. For more information on how to use the Methods feature in the application, see the Methods Section.
The BF Method combines two methods by splitting the Ultimate Losses into two components: Actual Losses and Expected Unreported (or Unpaid) Losses. As the years mature, more weight is given to the Actual Losses and Expected Losses (Loss Projection) gradually becomes less important. This method is a reasonable approach to estimating Ultimate Losses, especially for current or recently completed years, by smoothing the variance caused by the absence or presence of Large Claims.
The BF Method is useful for situations where the Actual Losses are not a good indicator of IBNR. This is often the case for low frequency but high severity lines of insurance. Another advantage of the BF Method is that it can be used even if there is not enough Historical Data. This method can be particularly useful when entering a new Line of Business. Additionally, the BF Method smooths the variance when there are random fluctuations or large claims at early maturities. This is useful for Long-Tailed Lines of Insurance such as medical malpractice or worker's compensation, particularly for the most immature years.
The Cape Cod Method, also known as the Stanard-Buhlmann Method, is used to calculate the Ultimate Losses in the Loss Reserves. Losses are projected through the Cape Cod Method by measuring both the Loss Exposure and Loss Development that occurs in a year. There are diverse volume measures that estimate the Loss Reserves for the Historical Accident Years. The Cape Cod Method, however, uses Volume Measures to project the Ultimate Losses for all Accident Years.
The Cape Cod Method is identified as a framework or an extension under the BF Method. It is a method that projects the Loss Development as well as patterns that signal the Ultimate Losses in Accident Years. The Cape Cod Method uses approaches that are more comprehensive than other methods, for example, many volume measures, and even the BF Method uses external information to calculate Loss Reserves for Accident Years. The Cape Cod Method, however, uses both internal and external information for the Ultimate Loss Calculations. The Cape Cod Method calculates the Loss Reverses by dividing the Loss to Date by Exposure, and then dividing it by the Ultimate Loss Development. The Cumulative Losses are also estimated by the Cape Cod Method.
A variety of trends and patterns are used in the general insurance industry to calculate some of the IFRS17 specific inputs. A few examples of these are Earning Patterns, Premium Receiving Patterns, Claims Payment Patterns, and so on. This section uses these Patterns, calculated from either the previous sections (e.g. Triangles) or via direct input, to calculate IFRS 17 specific inputs.
The output generated in the application is consumed by various processes and applications. Some of the examples are Reserving Exercise and Capital Modeling. A large portion of these outputs are also required for the IFRS17 computation and thus these work as an input for Oracle Insurance Accounting Analyzer (OFS IAA). The outputs that will be mapped to OFS IAA are:
· The Ultimate is calculated from each of the Loss Triangles, for example, Paid and Incurred Triangles. These losses form the expected losses for a particular Accident Year in the IFRS17 Application and will be mapped to the respective Accident Year.
· The IBNR is calculated using different methods, for example, the Chain Ladder Method. In case, the Incurred Loss has a component of IBNR, these are mapped based on the accident years in the IFRS17 Application.
· The Output is calculated from each variety of patterns, for example, Earned Premium. Depending on the variety of patterns, it can be mapped to the Underwriting Year (UWY) or Accident Year (AY). For example, the Earned Premium is mapped based on the UWY, and claim payment is mapped based on the AY.