Month: July 2023

Posted on

June 2023: Monte Carlo Simulation with Statistical Distribution Function, Data Table Sorting and Filtering, Improved Errors for HPM and more.

New

Monte Carlo Simulation

Slope now offers both a Random Number Generator and Statistical Distribution function in the formula editor so that they can be used to run Monte Carlo Simulation. Slope supports Probability, Cumulative and Inverse Cumulative function types and the following distributions: Beta, Binomial, Exponential, Gamma, Lognormal, Normal and Poisson.

The Statistical Distribution Function can be found by typing “Statistical Distribution” after opening the Formula Wheel or by clicking on the Math Functions icon.

After selecting your desired distribution type, you can use the formula wheel to provide a value, calculation or variable for the additional distribution parameters. When the Random Number Generator is used in combination with the Statistical Distribution function, users can run Monte Carlo simulations to predict a variety of outcomes.

Click here for more information on Monte Carlo simulation in SLOPE.

Enhancements

In App Sorting and Filtering on Data Table Columns

Clients can now sort and filter on data table columns to more easily review, sort and edit data without having to export their table to Excel. Columns with a data type of string, decimal, integer, boolean, date and drop down list can be sorted or filtered.

Sort order updates and ‘clear column’ changes can be saved by using the save button but data value filtering will not persist upon save.

Clicking the arrow next to the header label will allow you to filter the columns using a value or a condition or clear the data from the column completely.
Click on the header label for any data table column (that is not a decrement or data table column type) to sort in ascending order. Click a second time for descending order and a third time to remove the sorting and return the data to the original order.

Improved Error Information for High Performance Mode (HPM) Projections

High Performance Mode (HPM) now offers improved error reporting. HPM errors are found on the Run Logs tab and Projection Run Summary on the Workbooks tab. A new ‘Error Report – HPM’ tab on the Projection Run summary will show new columns to provide additional info for troubleshooting. New columns include: variable name that errored, time index that the error occurred, data table and table structure (if the variable was a data table read), portfolio name, product name, model point record # and current array value when applicable. Errors are also grouped by the variable where the error originated.

High Performance Mode projections will show the variable errors grouped by the variable that the error originates from. Expanding the variable will show all the variables impacted as well as relevant details on scenario number, time index, model point record number, etc.
High Performance Mode errors can be found on the ‘Error Report – HPM’ tab of the Projection Run Summary.

On the Debug View for HPM projections, ‘Projection Time’ has been added and shows the time period that the projection was processing when the result was calculated. The column formerly called ‘Time’ has been updated to ‘Time of Result’

Easily Filter Workbooks to show Monthly, Quarterly or Annual Results

Workbook data tables and visualizations can now be easily filtered to view monthly, quarterly or annual results using a ‘View By’ control. Instructions on how to setup the control and filter the data are available here.

The Income Statement and Balance Sheet workbooks (e.g. GAAP IS/BS) found in the Slope System workspace provide an example of a workbook with a View By control.

Changes

  • The Operator Control is now integrated into the end of a formula instead of inserted in between two operands to better align with the typical order of operations for creating a formula.
  • Non breaking spaces on Data Table Data and Index columns are now being converted to a standard space so they do not result in an error retrieving records at runtime.
  • The Bond Equivalent option for Yield Curve Rates will automatically be selected when rates are fetched from the Treasury for scenario generation.

Note: The change to always calculate variables that are non time indexed has been put on hold. This will allow clients an opportunity to update their models to prevent excessive errors on variables or specific model points for a variable that did not need to be calculated. Communication with timelines and model update recommendations will be sent via email to all clients well in advance of this change.

Fixes

  • Slope will now return a descriptive error message when there is an invalid nested data table.
  • The validation message for an empty table will now be cleared when importing a ‘File Only’ table over top of an empty table.
  • The correct time index is again being reported in errors when using a non-default data table.
  • A complex array aggregation configuration will no longer result in an unknown error at runtime.
  • Calculating a table formula for the first time within a goal seek chain will no longer result in an unknown error at runtime.