Problem 1. If a set of paired data gives the indication that the regression equation is of the form ya e, it is customary to estimate a and 3 by fitting the line log i = log + - log to the points {(log, log y): i = 1,2,...,n} by the methods of least squares, where log() is the common logarithm (base 10). (a) (12 pts) Use this technique to fit a power function of the form = &- to the following data on the unit cost of producing certain electronic components and the number of units produced: Lot size Unit cost x Y 50 $100 100 $50 $25 $9 $5 250 500 1000 (b) (5 pts) Use the result of part (a) to estimate the unit cost for a lot of 300 components.