JMPer Cable
A technical blog for JMP users of all levels, full of how-to's, tips and tricks, and detailed information on JMP features
JMP 19 includes several new HTML profilers and enhancements. The most notable additions to JMP 19 are all the new platforms that are now supported in interactive HTML and JMP Live. Want to try them? Download the attached script.
Distribution profilers
The Distribution platform allows continuous data to be fit using several methods and personalities. It includes both continuous fit methods (such as normal and Cauchy) and discrete fit methods (such as Poisson). Once a fitted distribution is created, distribution profilers can be enabled in the report.
Bivariate profilers
The Bivariate platform allows fits to be created to the report in an attempt to create a regression model to predict the data. There are a variety of methods and models that can be created from simple linear fit lines to polynomial fit curves and robust Cauchy curves. Once a fit is created, profilers can be added to the report and exported to interactive HTML and JMP Live.
Life Distribution profilers
The Life Distribution platform is a specialized platform in the reliability and survival capabilities of JMP. The Life Distribution platform creates multiple sets of profilers, including Distribution, Quantile, Hazard, and Density profilers. These individual cause profilers are now interactive in HTML and JMP Live. Note: The Comparison profilers will not be interactive at this point.
Fit Life by X profilers
The Fit Life by X platform is another specialized platform in the reliability and survival capabilities of JMP. Just like the Life Distribution platform, the Fit Life by X platform includes several profilers that are interactive in HTML and JMP Live: Distribution, Quantile, Hazard, and Density.
Model Screening profilers
The Model Screening platform is a comparison platform that fits many different predictive models and provides summaries. As long as the individual profilers are supported, the Model Screening profilers are all interactive in HTML and JMP Live. Currently, JMP supports the comparison of decision tree, bootstrap forest, boosted tree, K nearest neighbors, naive Bayes, neural, discriminant, fit least squares, logistic regression, generalized regression methods.
General profiler enhancements
In addition to the new platforms, there are a multitude of mathematical and statistical methods that are now supported in HTML and JMP Live. There are too many to list, however methods such as Cauchy distribution, lognormal distribution, etc., can now be added to the standalone profiler in the graph menu and can be exported to HTML and JMP Live.
JMP also supports HTML Profilers with more than 99 factors.
Neural profilers now support multiple Y’s and categorical Y’s in HTML and JMP Live.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.