A crack in a plastic compound affects the strength of the material. In order to investigate the relationship between the strength and the length of a crack, an experiment was conducted where cracked pieces of the plastic compound were subjected to increasing loads until the point at which the) broke apart. DS 12.8.5 has the data set of the breakage loads and the lengths of the cracks. Show that a linear regression model with breakage load as the dependent variable and crack length as the explanatory variable is unsatisfactory because the residuals are first positive, then negative, and then positive again. Show that a better fit can be obtained with the model
breakage load = γ0eγ1×(crack length)
Using this model, what is the expected breaking load when the crack length is 2.1?