fitting a distribution regarding sampling information (like batch information)
Jun 8, 2020 11:21 AM
| Last Modified: Jun 8, 2020 1:17 PM(304 views)
I am going to fit a distribution on my data regarding the batch information. So, I have one column for response, and one column for batch information. Now, my question is that, in "Fit Distribution" platform, where can I enter "Batch" information?
How can I let JMP know that information?
Does it matter at all? Or we should disregard batch information?
You can enter batch as a “By” variable in Distribution. From there, you can apply separate fits for each batch. Before doing this, you could use Fit-Y-by-X, with Batch as your x-variable and response as y, to check for differences between batches. If there are no significant differences, you can probably ignore the batch column and treat the response as though it is all from the same dataset. Does this make sense?