Home › Forums › General Forums › Tools & Templates › Regression Between Defect Rate and Per Unit $ Value

This topic contains 4 replies, has 5 voices, and was last updated by Hemanth 1 month, 3 weeks ago.

Viewing 5 posts - 1 through 5 (of 5 total)

- AuthorPosts
- September 4, 2018 at 12:36 am #706924
Hi Members,

I am trying to conduct analysis between a forecast defect rate and TCQ/Unit cost.

Can we conduct regression analysis between the above two.

Usually $ values are never used in regression.

Looking forward to thoughts from the forum.rgds

SamSeptember 4, 2018 at 8:58 am #706931Yes you can. There are no restrictions on the distributions of the X’s or the Y’s in a regression analysis. As for “never” using dollar values in a regression equation – how would you propose to identify an optimum combination of ingredients for, say a plastic compound, such that the dollar cost per pound was minimized other than using the X’s to predict the final dollar cost. In short – perhaps someone has told you this is never done but whomever offered that opinion is wrong.

September 11, 2018 at 11:31 am #706965@rbutler Robert “whomever offered that opinion is wrong.” Very nice. You are always to tactful (generally). Had to smile when I saw this one.

September 16, 2018 at 2:13 am #707008usually $ values not used in regression? ….why not

anyways, I would expect COQ to have at least be part of a multiple regression with parts or $ since the basic COQ would involve them in the calculation. Of course, maybe the COQ could be “recorded wrong” or have bad assumptions.

September 16, 2018 at 5:17 am #707010Technically speaking, you can use any parameter in regression, dollar value is no exception. However, the question is how useful is the $ value when you run the regression for processes.

Remember, $ value is a derived parameter based on consumption and unit rate. Hence, its always useful to apply some caution with $ figures as they can get confounded. - AuthorPosts

Viewing 5 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic.