Additional functionality of multiple non-linear regression analysis
A description of the enhancement required:
To move beyond the statistical forecasting model from Anaplan, additional functionality of multiple non-linear regression analysis and ability to derive data set screening with statistical significance p-values, r-squared coefficients, and more advanced modeling as is being requested by end users. Such functionality can currently be accomplished using R programming language and is now a supplementary external process we’d like to house within Anaplan. Taken a step further, this would get into AI modeling and predictive analytics not unlike what was shown at Hub this year.
An example of the enhancement:
Addition of ARIMA (Autoregressive Integrated Moving Average) and Neural Network Models for time-series forecasting. More functionality on being able to manipulate subsets and inflection points within forecasted values. Being able to set forecast windows. Alternatively, having some sort of ‘R’ connector much like the Excel connector.
A story for why the enhancement is required (specifically, how would it aid the business process):
Our forecasting team initially adopted the statistical modeling models we developed for the team using Anaplan. They require the ability to analyze standard deviations, more advanced controls of ‘fine tuning’ algorithms, prediction windows, multi-variables and statistical significance calculations (things that are possible in R). Understanding this request requires more advanced stat modeling.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.