price determination prediction

Its been 2 years since my last statistics class. I did fairly well but the material isn't so fresh since I haven't had to use it.
I am writting a sales contract and want ideas / comments on price determination.

Say for example your looking to sell honey and market prices are published daily.
Published prices generally are correct but are mostly 1-2 dollars LESS than those of the daily market report.

Consider you need to specify a "fair" market predictor to nail the price determination section beyond vague terms/undefined methods to protect your interests.

Here goes a couple ideas that initially pop-ed up.

1. If I were to have access to the potential clients historic purchase prices. I can search the corresponding market prices for those purchase dates, put them in a spreadsheet and continuously predict a sales price (y) based on market pice (x) with a linear regression. + Since actually determined price won't be based on prediction this data will be manually entered as distribution progresses.

*this predicted ybar isn't an interval.

Does anyone have the equation for making a confidence interval for a linear regression?
Can anyone comment on a proper & legally plausible bound contruct considering the market price is typically 1-2 dollars LESS than the market prices?

2. If I didn't have access to the potential clients historic purchase prices would simply stating the price must be within x% of published price be my best way to specify whats acceptable? It seems like you wouldn't need statistics for this.

3. Any other comments or suggestions?