Beginner's question: Is profit/loss data always ratio in nature? How to treat it?

[MrPink]

Hello, I'm afraid this is another beginner's question. I've only recently grasped the fundamentals of linear and multiple linear regression and I still have the following nagging doubt.

I'd like to analyse some sales data for the purpose of forecasting future performance. My dependent variable (Y) is 'profit/loss'. My question stems from the fact that the historical values for this variable are either a positive number (ranging from 0 to 1000) or a fixed negative value of -100. It's my understanding that a multiple regression model here would produce varying negative (and positive) values for Y, and this is not my issue. What I'd like to know is, are there any other implications of using this sort of input in a regression model? Or can it be treated in the same way as any ratio type data? Thank you.

 

[Dason]

I don't understand what your historical values are representing. Can you elaborate a little bit more.

 

[MrPink]

Hello, the 'profit/loss' values are simply sales figures for individual items. An item may be sold for any amount of profit but the wholesale price to the seller of each item is the same, hence the same fixed loss amount for any unsold items. That is the variable I would like to forecast. There are several independent variables for each attempted sale of an item, which of course vary (such as sale location, duration of item availability, funds spent on advertising, etc.)

 

[MrPink]

To clarify further, a sample of the data I'd like to use might look something like the following (note the fixed negative value of -100 for the dependent variable in a couple of instances):

X1	 X2	 X3	 Y
893	 4.72	 -65	 23
980	 9.14	 88	 55
456	 2.30	 37	 201
412	 5.08	 13	 -100
-191	 1.96	 64	 13
902	 -3.83	 41	 -100
893	 7.35	 75	 321

Perhaps it sounds silly but I'm wondering whether the fixed negative values might somehow 'skew' a regression model's output. As I say, I'm not trying to replicate the fixed -100 value for the losses, only trying to get to accurate averages such that I may accurately determine the profitability of an item's listing for sale given certain quantifiable conditions. Make sense?