For insurance companies to remain profitable they need to collect more in yearly premiums than they # spend on the medical care of their beneficiaries. # # Appropriate medical expenses are difficult to estimate because the mostly costly conditions are rare # and seemingly random. Still, some conditions are more prevalent for certain segments of the population. # For instance, lung cancer is more likely among smokers; heart-disease is more likely among the obese. # # Use patient data named insurance to estimate the average medical care expense for such # population segments. The data contains data on 1338 beneficiares currently enrolled in the insturance plan. # It contains information on total medical expenses charged to the plan (charges).

# ############################################################################

#1. Get the names of the variables.

#2. Examine correlation among the explanatory variables - age, bmi, children - and charges, using cor() # What two variables have the strongest relationship?

#3. Using age, children, bmi, and sex, estimate a linear regression model for charges. Name it model1.

#4. How much are additional (average) charges for each year that age increases?

#5. How much are additional charges for each additional child?

#6. How much are additional charges for male than female ?

# 7. What does the *** and * mean in the summary near the p-value?

# 8. Is the adjusted R2 strong in this model? What does the adjusted R2 tell you?

# 9. Predict charges for the individuals for age 30, sex is male, bmi is 33.77, children is 0.

# 10. find the mean of residuals of the model. Is it zero, why do we want mean errors to equal zero?