# fitting a curve to a scatter plot

#### mpz

##### New Member
It's been over 9 yrs since my last statistics course so maybe someone can help me out.

Given (x, y) values where x = number of days passed (but also let it be price in dollars) and y = number of widgets sold, find lowest most effective price to purchase a widget (hope that makes sense).

(0, 10)
(100, 20)
(200, 30)
(300, 10)
(400, 40)
(500, 25)
(600, 15)
(700, 30)
(800, 40)
(900, 50)
Code:
(#)
50+                                   *
|
40+               *               *
|
30+       *                   *
|                   *
20+   *
|                       *
10+           *
|
+---+---+---+---+---+---+---+---+---+
100 200 300 400 500 600 700 800 900 (t)
9.1 9.2 9.3 9.4 9.5 9.6 9.7 9.8 9.9 (\$)
I know I can use line of best fit to predict future values for trend but how do I find the lowest most effective price based on data for the time frame?

I think that if I somehow fit a bell shaped curve (distribution curve) to above data I can calculate probability of any y-value and thus finding my lowest most effective price. I'm at a loss as to how to do that.

I've been trying to figure this out for two days. Any help you can offer is greatly appreciated.

Cheers