Automatically generating reports in Degradation Data Analysis platform?

AR80615

Hi,

When analyzing degradation data using the Degradation Data Analysis platform, I want to screen various models in order to select the best one based on AICc. To do this I am manually applying the different distributions and transformations for a given path definition, and generating reports for each combination by clicking the "Generate Report" button. Is there a way to have JMP automatically generate reports for all the combinations and add them to the Model List? A bonus would be to have JMP perform this while also excluding transformations that are deemed inappropiate or generates missing values (i.e. applying log-tranformation to the time factor when it contains zero).

JMP version 17.2

Thanks in advance.

/JF

 

 

bilde.png

jthi

You might have to do this by scripting. Below is one option (not very robust script, but can give some ideas about what can be done)

Names Default To Here(1);
dt = Open("$SAMPLE_DATA/Reliability/Adhesive Bond.jmp");
obj = dt << Destructive Degradation(
	Y(:Strength),
	Time(:Weeks),
	X(:Degrees),
	Censor(:Censor),
	Censor Code("Right"),
	Control("Log10", "Sqrt", "Normal", "Individual Path with Intercept")
);

rep = Report(obj);

cb = (rep << XPath("//PanelBox[text() = 'Distribution']/ComboBox"))[1];
rb_transformations = (rep << XPath("//PanelBox[text() = 'Transformation']//RadioBox"));

cb_options = cb << get items;
rb_tr = rb_transformations << get items;
btn = (rep << XPath("//ButtonBox[text() = 'Generate Report']"))[1];
tb1 = rep[TextBox(3)];
tb2 = rep[TextBox(4)];

For Each({option, idx}, cb_options,
	cb << Set(idx);
	For Each({op1, idx1}, rb_tr[1],
		rb_transformations[1] << Set(idx1);
		For Each({op2, idx2}, rb_tr[2],
			rb_transformations[2] << Set(idx2);
			visibilities = {tb1, tb2} << get visibility;
			visibs = Loc(visibilities, "Visible");
			If(N Items(visibs) == 0,
				btn << Click(1);
			,
				skipped_txt = Eval insert("\[Skipped combination ^option^+^op1^+^op2^ due to ^Concat Items(({tb1, tb2} << get text)[Matrix(visibs)], ", ")^]\");
				Write("\!N", skipped_txt);		
			);
			wait(0);
		);
	);
);

Write();