# Price elasticity for n data points

So let's say I have 100 data points which contains the price and sales of a product. Just to start I assume that the relationship is of the form Q = a - b*P where P is the price and Q is the quantity. Now I know that if i run a regression model using this equation then b*(P/Q) will give me point elasticity for each consecutive data point. My question is that I will have 99 elasticity values for the 100 data points, so what do I call as the overall elasticity, should I average the 99 elasticity values and call that as the final elasticity value?

I am new with the concept of elasticity, so pardon me if I am sounding horribly wrong with my logic.

• Elasticity is b no? the ratio between change in q to change in p. Since you are using a linear regression - with only one coefficient describing the relation between price and demand) - You get one elasticity value for P's in your dataset. – yoav_aaa Aug 21 at 6:39
• Elasticity is b when I use a log log model i.e. log of quantity and price and then run the regression model. – Brohn Aug 21 at 7:37
• Take a look at arc elasticity(average on point wise elasticity) - en.wikipedia.org/wiki/Arc_elasticity – yoav_aaa Aug 21 at 7:44