In a data set with n = 50 observations, you try fitting two models. The first model is a simple linear model (m 1 = 1) resulting in SSE 1 = 932. The second model is a cubic polynomial (m 2 = 3)...

In a data set with n = 50 observations, you try fitting two models. The first model is a simple linear model (m₁
= 1) resulting in SSE₁
= 932. The second model is a cubic polynomial (m₂
= 3) resulting in SSE₂
= 901. Did the second model fit significantly better than the first model? Give the formal hypothesis that corresponds to this question, and show the construction of the appropriate test statistic. Use α = 0:01.