(a) Examine the output in EXHIBIT III below. Use relevant values from the output as well as one of the words: 'increase' or 'decrease' to fill in the blanks in the gray shaded area below. % of the variation in price of wine can be explained by the relation between alcohol measured and price of wine, namely that an increase of one unit of alcohol measured results in a(n) of $ in price of wine. EXHIBIT III Bivariate Fit of Price By Alcohol measured 14 12 Price 2 11.5 12 125 13 13.5 14 14.5 Alcohol measured -Linear Fit - Bivariate Normal Ellipse P=0.950 Linear Fit Price = -28.34218 + 2.6985005*Alcohol measured Summary of Fit Square 0 256613 RSquare Adj 0.243571 Root Mean Square Error 2.564406 Mean of Response 6.774915 Observations (or Sum Wgts) 59 Parameter Estimates Term Estimate Std Error t Ratio Prob>|t/ Intercept -28.34218 7.923834 -3.58 0.0007 * Alcohol measured 2.6985005 0.60835 4.44