Email this page    Print this page

JMP Pro Software

JMP Pro includes powerful new capabilities for advanced analytics users who need more modern methods with greater automation to create robust predictive models

JMP Pro provides all the superior visual data access and manipulation, interactivity, comprehensive analyses, and extensible capabilities that are the hallmarks of JMP, then adds modern predictive modeling, cross-validation, exact measures of association, one-click bootstrapping and model comparison features. All of this comes in the in-memory, desktop environment familiar to current JMP users

JMP® Pro, Version 10 Features

Visual discovery

Graph Builder is the interactive drag-and-drop tool for creating a graphical analysis. This is among the many tools available for data visualization in JMP Pro.

Build predictive models that generalize well

Use cross-validation and build boosted neural network models with automatically generated SAS score code

Interactive Data Mining:-

Build boosted decision-tree models and generate SAS score code.

Compare predictions across many modeling techniques

The model comparison platform lets you compare fits across multiple fit predictions

Assign measures of precision to model predictions

Bootstrap distributions of parameters of a non-linear model.

Exact tests

Here, the Cochran-Armitage test is used to look for increasing incidence of drug effect with increasing dose. Note the difference between the asymptotic p-values and those from the exact permutation test.

Key features exclusive to JMP® Pro, Version 10:

Validation column role in many modeling platforms
Bootstrapping for most statistics in JMP reports

Fit Model platform

Partial least squares implementation allows specification of model terms
Stepwise regression uses train, validate and test methodology

Partition platform

Bootstrap forest and boosted tree techniques
Uses train, validate and test methodology

Model Comparison

Compares the fit of different models
Provides measures of fit, ROC, diagnostic plots and profilers

Contingency platform

Exact measures of association

Oneway platform

Nonparametric exact tests

Neural platform

Automated handling of missing data
Automatic selection of the number of hidden units using boosting
Automated transformation of input variables
Outlier-resistant loss function
Ability to fit both one- and two-layer neural nets
Ability to choose from three activation functions
Ability to save transformed covariates
Ability to save randomly generated cross-validation columns
Uses train, validate and test methodology

PLS platform

Choice of k-fold or Holdback validation methods

PCA Platform

Variable clustering option for predictor variable reduction prior to modeling

In This Section

Interested? contact us

Easy ways to contact our sales representative.
USA : +1.732.548.9268
UK : +44.20.3371.9221
INDIA : +91.124.4017660
Email us Email us

Related Information

Product Engineering Solutions
Product engineering solutions has become the requirement for all companies.
Software Development Company
Software development companies in India have grown largely over the past couple of years with major IT giants establishing their base.

Technical Support

Get free technical support Request for demo cd

Our Alliance