MicroStrategy ONE

Advanced Options

This step provides the ability to set analysis-specific options which dictate how the selected training algorithm should behave. There is also the option to configure the behavior of each input variable. The variable settings affect how the input variables are processed by the training algorithm and, later, by the scoring engine.

The Advanced Options are not available for time series and association rules analysis.

Variable Reduction Settings (linear and exponential regression only)

One area of regression analysis deals with variable reduction. This pertains to methods of removing insignificant independent inputs from the resulting model. Input significance is determined by Variable Importance, which can be adjusted by the user by means of a slider. The range of possible values is zero (no predictive value) to one (perfect predictor).

MicroStrategy supports three types of variable reduction:

  • Single pass, which performs a single pass over the variables, calculating alpha values for each, and eliminating those variables with alpha values less than the user-specified alpha.

  • Forward, which starts with a null model (with no variables) and in every iteration adds the variable which is the most significant (that is, the lowest alpha value). At each step, new alpha values are calculated, and this process continues until no variables can be added to the model whose significance is less than or equal to the user-specified alpha.

  • Backward, which starts with a full model (with all of the variables) and in every iteration removes the variable which is the least significant (that is, the highest alpha value). At each step, new alpha values are calculated, and this process continues until no variables can be eliminated from the model whose significance is greater than the user-specified alpha.

Do one of the following: