Question

In an experiment to study the effect of temperature (x) on the yield of a chemical reaction (y), 30 experimental runs were conducted. The level of temperature was carefully controlled at each of five levels, coded as x = -2, -1, 0, 1, 2. Two catalysts were used. For each catalyst three runs were taken at each level of temperature, and the yield was measured. The model y = beta_0 + beta_1x + beta_2x^2 + beta_3z + epsilon, epsilon ~ N(0, sigma^2) was considered, where z = 0 for catalyst 1 and z = 1 for catalyst 2. a. Carefully interpret the parameter beta_3 in this model. b. The model was fit to the data and the output is summarized below. The residual sum of squares is 25.05, and Is there any evidence of a difference in the two catalysts? Find a 95% confidence interval for beta_2. c. We also know that (X'X)^-1 = [0.114 0 -0.023 -0.067 0 0.017 0 0 -0.023 0 0.012 0 -0.067 0 0 0.133] i. Explain why ^beta_1and ^beta_3 are independent random variables. ii. Find a 95% confidence interval for the expected yield when the standard temperature (x = 0) and catalyst 2 are used. iii. Find a 95% prediction interval for the yield of a new experiment run under standard temperature (x = 0) and with catalyst 2.

Answer 1

=5

Explanation:

Given that an experimenter is studying the effects of temperture, pressure, and type of catalyst on yield from a certain chemical reaction. Three different temperatures, four different pressures, and five different catalysts are under consideration.

a) Experimental runs possible if use of single temperature, pressure and catalyst is there = no of temperatures x no of pressures x no of catalysts

= b) Here pressure and temperature have no choice as lowest is selected.

no of methods = no of catalysts x 1 x1

= 5