If your talking about test for equivalence of means allowing unqual variances, I believe the can be done, but "indirectly". Equivalence test of means (as, say 95% level) is given by (90%) confidence interval for difference in means. Believe the t-test in the Fit Y by X platform gives that confidence interval allowing unequal variances,
Indeed, I've had quite a bit of success with the Graph Builder, ... as well as Control Chart Builder and Local Data Filter. Many Excel users love the "pivot table feel" with power of graphics.
Another sucess I've had at showning value is replacing hours of manal copy-paste and cell formulating in Excel with seconds via a JMP script.
The Help and Books are great also!
That's true! Also the syntax for conditional logic statements in JMP column formulas is WAY more intuitive than in Excel. I can write conditional statements with very complex logic using JMPs graphical display dialogue (with JSL embedded underneath) and in this way confidently create new column formulas to help me manipulate and gain greater insights from my data.
Good stuff Mark. I think this is a challenging thing for some of the larger companies when they get their hands on JMP. The challenge may be one of communication. Folks come to them and say "hey, JMP is a great way to analytics better than what you are doing in Excel", which is very true, mostly, with the important caveats that JMP is not a spreadsheet and there are certain things that are much better to do in a spreadsheet.
But what the company may hear is "I can do everything I did in Excel better in JMP". I believe that this can result in some serious deployment heartburn as folks are tasked with transitioning things out of Excel into JMP when Excel still sits as the more fit-for-purpose tool.
One example from my own experience is discreet event simulation. Now if I'm doing any kind of heavy duty work there is more appropriate software for it, but for low-end back of the envelope queuing models I go straight to a spreadsheet program.
Monte-carlos in general I find to be somewhat evenly split. There are a lot of places where it's small enough I don't need to go to Python, but it's complex enough that its much easier to do it in a spreadsheet software than in JMP.
Linear programming as well. If it's a lightweight model I'm gonna run that in spreadsheet software. I'm not even certain JMP has that capability.
The best way to summarize the difference in my opinion is this: JMP is one of the best tools out there for doing a specific set of tasks, but it is not intended for everything. Excel is is a tool that can do entry level work on almost any task. Or, maybe: JMP can do a few things excellently. Excel can do everything poorly.
Maybe JMP is a spreadsheet? Or at least it can be! What I've been doing of late is just using the cell shading (dark blue) to "hide" duplicated calculations that I am performing which automatically duplicate in all the rows of a given column. Works pretty well! Only thing I wish was that it was even easier to access column formulas than it is currently. Maybe instead of a couple clicks it could be accessed and you could be taken there directly with a single right click or some other shortcut. The handy thing about Excel is the ability to readily enter column formulas. JMP you can do the same once you get in there to the formula editor and you can even use the JSL, but it takes some time to even access the formula editor, especially if you're doing this for multiple columns in one data table.