How do I create a correlation or fit model to identify power consumption increase?Hi, I am an absolute beginner in JMP and I am interested to know an accurate representation of what is the extra power that my motor will consume if the filter dP is at 1500 psi vs 2000 psi.
I did a multivariate analysis and using my process knowledge I am able to narrow down the parameters that will affect the power consumption. I did a preliminary test using fit model and able to create an equation with a pretty solid Rsquared however I am not sure if this is the right track. Should I then use the prediction profiler and set my dP at 1500 vs 2000 psi and look at the resulting power comparing those two?
Basic Data Analysis and ModelingRe: How do I create a correlation or fit model to identify power consumption increase?I notice that you colored the points by year but did not include year in your model - I would add year as an effect. But I believe your conjecture about how to proceed is appropriate - you can use the profiler to see the estimated power along with confidence intervals from your model. I would only add that I think there might be a bit more accurate model lurking if you explore the behavior over time. Month seems to entail both power and dP varying (with month as a nominal variable) and produces a slightly better model. I don't understand your motor and the process at all, so that expertise probably would help understand how/why things might change throughout the year. But I didn't see any interaction between the year and the effects of dP on power.
