]]>All FAQsWed, 17 Feb 2016 02:56:41 -0300AutoDiscovery of Predictors in SPM®
https://www.salford-systems.com/?id=350:autodiscovery-of-predictors-in-spm
https://www.salford-systems.com/?id=350:autodiscovery-of-predictors-in-spmAutodiscovery leverages the stability advantages of multiple trees to rank variables for importance and thus select a subset of predictors for modeling. In SPM® v8.2 and earlier Autodiscovery runs a very simple training data only TreeNet model growing out to 200 trees. The variable importance ranking generated from this model is then used to reduce the list of all available predictors down to the top performing predictors in this background model. Autodiscovery is fast and easy, as there are no control parameters to set, but it is just a mechanism for quickly testing whether a substantial refinement in the number of predictors can improve model performance.
]]>AutoDiscovery of Predictors in SPM®Wed, 17 Feb 2016 03:03:56 -0300Can We Obtain Dependency Plots for Single CART® Trees?
https://www.salford-systems.com/?id=381:can-we-obtain-dependency-plots-for-single-cart-trees
https://www.salford-systems.com/?id=381:can-we-obtain-dependency-plots-for-single-cart-treesThe short answer is YES such plots can be generated. Historically, we concluded that such graphs would normally not be that interesting as they would frequently be single step functions reflecting the fact that individual variables often appear only once or twice in a tree. Also, such graphs would not properly reflect the effect of a varible across most of its range of values. Thus, as of SPM® 7.0 CART® does not offer such plots. However, we can see what such plots would look like by using TreeNet® to grow a one-tree model. To do this, just set up a normal model, choose the TreeNet analysis method, and set the number of trees to be grown to 1 (see green arrow below).
]]>Can We Obtain Dependency Plots for Single CART® Trees?Wed, 17 Feb 2016 03:46:29 -0300CART® and Large Datasets
https://www.salford-systems.com/?id=371:cart-and-large-datasets
https://www.salford-systems.com/?id=371:cart-and-large-datasetsCART is capable of determining the number of records in your data sets, and uses this information to predict the memory and workspace requirements for trees that you build. Also, CART will read your entire data set each time a tree is built. At times these actions may be problematic, especially if you have enormous data sets.
]]>CART® and Large DatasetsWed, 17 Feb 2016 03:39:12 -0300CART® Feature Matrix
https://www.salford-systems.com/?id=345:cart-60-feature-matrix
https://www.salford-systems.com/?id=345:cart-60-feature-matrixFeature Matrix and download PDF
]]>CART® Feature MatrixWed, 17 Feb 2016 02:59:18 -0300How are nominal (ordered) predictors and rank related?
https://www.salford-systems.com/?id=375:how-are-nominal-ordered-predictors-and-rank-related
https://www.salford-systems.com/?id=375:how-are-nominal-ordered-predictors-and-rank-relatedOne of the strengths of CART® is that, for ordered predictors, the only information CART uses are the rank orders of the data – not the actual value of the data. In other words, if you replace a predictor with its rank order, the CART tree will be unchanged.
]]>How are nominal (ordered) predictors and rank related?Wed, 17 Feb 2016 03:41:52 -0300How can MARS® models be implemented for predictive purposes?
https://www.salford-systems.com/?id=388:how-can-mars-models-be-implemented-for-predictive-purposes
https://www.salford-systems.com/?id=388:how-can-mars-models-be-implemented-for-predictive-purposesA MARS® predictive model can be implemented in two ways. First, new databases can be scored directly by identifying the MARS model and the data to be scored. MARS will perform all the required data transformations and calculations automatically and output the predicted scores. Second, the MARS predictive equation can be exported as ready-to-run C and SAS®-compatible code that can be deployed in the user's application framework.
]]>How can MARS® models be implemented for predictive purposes?Wed, 17 Feb 2016 03:52:03 -0300How does MARS® compare with neural nets?
https://www.salford-systems.com/?id=389:how-does-mars-compare-with-neural-nets
https://www.salford-systems.com/?id=389:how-does-mars-compare-with-neural-netsMARS® is not a black box. It is faster, more interpretable, and more accurate than neural nets.
]]>How does MARS® compare with neural nets?Wed, 17 Feb 2016 03:52:45 -0300How does MARS® construct its models?
https://www.salford-systems.com/?id=384:how-does-mars-construct-its-models
https://www.salford-systems.com/?id=384:how-does-mars-construct-its-modelsMARS® starts from the premise that most relevant variables affect the outcome in a complex way. Therefore, when MARS considers whether to add a variable, it simultaneously searches for appropriate break points – knots. Models are constructed in a two-phase procedure. Phase I tests variables and potential knots, resulting in an overfit model. Phase II eliminates redundant factors and components that do not stand up to testing.
]]>How does MARS® construct its models?Wed, 17 Feb 2016 03:49:31 -0300How does MARS® differ from conventional regression?
https://www.salford-systems.com/?id=383:how-does-mars-differ-from-conventional-regression
https://www.salford-systems.com/?id=383:how-does-mars-differ-from-conventional-regressionConventional regression models typically fit straight lines to data. MARS® approaches model construction more flexibly, allowing for bends, thresholds, and other departures from straight-line methods. MARS builds its model by piecing together a series of straight lines with each allowed its own slope. This permits MARS to trace out any pattern detected in the data.
]]>How does MARS® differ from conventional regression?Wed, 17 Feb 2016 03:48:38 -0300