Release History

This release is focused on improving the user experience and system performance around large amounts of data, because no one likes waiting for things to load!

NEW

The file manager and related file upload functionality has been completely overhauled. This change results in faster upload times, an increased file size limit, and new features such as drag & drop, pausing, resuming, and retrying.

IMPROVEMENTS

• Improved data table import performance.
• Data tables up to 100,000 rows can now be imported.
• Significant performance improvements around loading long projection run logs, particularly if a large number of errors were logged.
• The data table editor is now limited to editing the first 1000 rows, to improve the speed of the editor. Users can edit more rows by exporting the table, editing, and re-importing over top of the existing table.

IMPROVEMENTS

• We’ve upgraded some underlying technologies in the calculation engine resulting in some minor performance improvements, increased security, and improved stability.
• When copying a model, the various settings on projection templates are now maintained.

FIXES

• Fixed an issue in the formula builder where certain actions caused multiple formulas to be active at one time, resulting in difficulty interacting with other elements in the formula.
• Fixed an issue causing certain projections to use more memory than necessary when executing within the calculation engine.

Introducing… nested data tables! Now, you can create tables that reference other tables (similar to how nested mortality tables work). This allows for drastically simplified management of certain information.

NEW

Nested Data Tables

• First, a new column type of “Data Table” is available when configuring table structure columns.

• These columns can then be configured to point to individual data tables within the assumption management module using the table editor.

• Now, a new type of variable (product, portfolio, and company) return type is available. Specify a variable with a return type of “Data Table” to make use of a nested table structure. Within this variable, you can return the column value you created in the previous step.

• The last step is to specify the data table variable within a table reference inside the formula builder (the default option will still use the projection table mapping). This allows the calculation engine to determine the data table at run time using the configured nested data tables.

Reporting Module

• The default table visualization has been updated. The original table visualization is still available as “Table (Legacy)”. This feature will be available as of Nov 21st.

IMPROVEMENTS

• Security is of utmost importance here at Slope. As such, we’ve rolled out several back-end changes this release for even further increased security.
• Display precision on variables at all levels is now only shown if the variable is of type ‘Decimal’.
• Models are now sorted by name in the Compare Models and Merge Models interfaces.

We’ve rolled out our rebranded look to the SLOPE platform! We hope you enjoy this modernization to our brand.

As part of our ongoing effort to better support our clients, we’ve also rolled out a live chat feature on the help page. Users can navigate to this page from the User menu in the upper right-hand corner, or by going straight to help.slopesoftware.com. The icon pictured below will appear in the lower right-hand corner and allow users to chat with a SLOPE representative in real-time.

The MetLife Digital Accelerator is still keeping us very busy, but keep your eyes out for more feature releases towards the end of the program. That being said, here are a few fixes for bugs that were brought to our attention recently. We hope these make your lives easier!

FIXES

• Fixed a display issue with the “Variable Tags” field on projections and projection templates causing tags that were no longer on variables to be hidden from view, but the tag was still filtering variables during execution.
• Mortality table dropdown selection now works properly in Firefox when editing data tables.
• Fixed errors caused downstream by changing a table column type from mortality table to something else (such as decimal, integer, etc).

We’ve been hard at work at the MetLife Digital Accelerator, but in the meantime here are a few features and bug fixes that have been actively requested.

NEW

• Variables within an array variable calculation chain can now be changed to have a formula per array value. This allows more flexibility and different logic in different parts of the dependency chain.

FIXES

• Fixed an issue where formulas were sometimes duplicated when moved via drag & drop outside of a grouping.
• Fixed an issue with nested array formulas that sometimes caused unintuitive behavior around which array value was used during calculation.

NEW

• Individual products can now be selected during copy and export of a model to give users more fine grained control of what is copied or exported.
• Arrays and array values can now be merged between models.

IMPROVEMENTS

• Portfolio cash references (before/after investment) are now marked with “t-1” to make it clear that the value referenced is from the end of the previous time period.

FIXES

• Fixed an issue where if a product variable was set as an array variable the product always ran in dynamic calculation mode even if it could have been run in static mode, therefore hurting performance.
• Fixed some issues around copying, merging, and importing models containing array variables.

IMPROVEMENTS

• The current array value is now logged in calculation engine error messages, when calculating array variables.
• Performance has been improved when calculating goal seek functions within array variables.
• Models are now sorted by name across the site.
• Minor performance improvements during page load in the Analyze module.

FIXES

• Input table formulas within aggregate formulas now properly show validation messages when indexes are missing.
• Fixed an error that sometimes occurred after merging a variable with an input table formula referenced within an aggregate formula.

NEW

• We’ve added the concept of data table index “read methods”, which allows for the configuration of how data table rows are read. In this release, we’ve added the “Less Than or Equal” operator, which will match an index during calculation if the index is less than or equal to the value provided.

IMPROVEMENTS

• Drag handles have been added to the headers on tokens within the formula builder to make it more apparent that dragging is supported and to give an alternative to clicking and dragging on the main part of the formula.

• Self referencing variables’ self dependencies are now handled automatically when merging.

FIXES

• The link to input tables from the formula builder has been fixed.
• Empty groupings on the formula builder can no longer be “ungrouped”, creating an invalid formula.
• Fixed an interface issue where some items on the page were still able to be interacted with after activating a menu on the formula builder.
• Data tables are now appropriately updated when merging in input table column changes.
• Fixed a merging issue that caused some variables to have invalid references after merging a new portfolio/product into a model.
• Fixed a model copy error that occurred on some models.

Expanding on our last release, this release focuses on some improvements to the newly available array variables feature, as well as a few other improvements.

NEW

• A new formula type is available allowing users to reference an array index value directly.

• A new system variable is available which allows users to reference the current array index value being calculated during runtime.

IMPROVEMENTS

• Array values are now displayed as their own nodes in the relationship view for array variables, to better display the variable dependencies. Formulas for each index value can be viewed, as well as their dependencies.

• Formulas for array variables can now be copied to specific index values in addition to being able to be applied to all index values at once.

• Improved performance around saving array variable changes at the bottom of a large dependency chain.
• Performance around model import has been improved.
• Minor calculation engine performance improvements.