Would model m1 be sufficient for modeling this data? Justify your answer by providing a statistical evidence...

The data in the following table are from a famous study conducted by Sir Richard Doll and colleagues. In 1951, all British doctors were sent a brief questionnaire about whether they smoked tobacco. Since then information about their deaths has been collected. The table shows the numbers of deaths from coronary heart disease among male doctors 10 years after the survey. It also shows the total number of person-years of observation at the time of the analysis Woud agesq be a variable we should include in the model? Justify your answer by developing a suitable statistical test. Among those four models, which one is the best in your opinion? Support your choice by using at least two approaches which must include the likelihood ratio test. From your best model, interpret the coefficient of smoke, and explain how the smoke status affect the deaths. we have the following fitted values (5 points) 25.584734 106.811960 208.198646 182.827893 102.576767 3.414801 11.541629 24.743377 20.229155 31.071038 Perform a formal test to verify whether the fitted values are good enough.