Practical Computing Exercise for Week 1 :Reworking of the baby weights example Solutions

Aims of this practical exercise

In this exercise you will:

  • rework the example given in ELMER
  • get some practice using R markdown

Before you undertake this exercise…

You need to have installed R, RStudio, and the necessary packages for the course, including the ELMER package. See how to get set up for this course

data(BabyWeights, package="ELMER")
str(BabyWeights)
'data.frame':   24 obs. of  3 variables:
 $ Age   : int  40 36 40 38 42 39 40 37 36 38 ...
 $ Weight: int  3317 2729 2935 2754 3210 2817 3126 2539 2412 2991 ...
 $ Gender: Factor w/ 2 levels "Female","Male": 1 1 1 1 1 1 1 1 1 1 ...

Fit all possible models

As per the example in ELMER, you need to fit five models. Make sure you label them carefully to match ELMER.

Baby.lm1 = lm(Weight~1, data=BabyWeights)
Baby.lm2 = lm(Weight~Gender, data=BabyWeights)
Baby.lm3 = lm(Weight~Age, data=BabyWeights)
Baby.lm4 = lm(Weight~Age+Gender, data=BabyWeights)
coef(Baby.lm4) # needed later
(Intercept)         Age  GenderMale 
 -1773.3218    120.8943    163.0393 
Baby.lm5 = lm(Weight~Age*Gender, data=BabyWeights)
NewBabies = data.frame(expand.grid(Age=c(36, 38, 40,42), Gender=c("Female", "Male"))) 
NewBabies = NewBabies |> mutate(Preds1=predict(Baby.lm1, NewBabies), Preds2=predict(Baby.lm2, NewBabies),
Preds3=predict(Baby.lm3, NewBabies), Preds4=predict(Baby.lm4, NewBabies), 
Preds5=predict(Baby.lm5, NewBabies))

Make the graphs

N.B. You should try to do this using ggplot() if you can.

OverallSum = BabyWeights |> summarise(Weight=mean(Weight))
OverallSum
    Weight
1 2967.667
FemaleSum = BabyWeights |> filter(Gender=="Female") |> summarise(Weight=mean(Weight))
FemaleSum
    Weight
1 2911.333
MaleSum = BabyWeights |> filter(Gender=="Male") |> summarise(Weight=mean(Weight))
MaleSum 
  Weight
1   3024

Starting with:

Graph0 = BabyWeights |> ggplot(aes(y=Weight,x=Age)) + geom_point(aes(col=Gender))
Graph0+geom_hline(yintercept=OverallSum$Weight)

Graph0+geom_hline(yintercept=FemaleSum$Weight, col="red") + geom_hline(yintercept=MaleSum$Weight, col="blue")

Graph0+geom_smooth(method="lm", se=FALSE)
`geom_smooth()` using formula 'y ~ x'

Graph0+geom_abline(slope=coef(Baby.lm4)[2], intercept=coef(Baby.lm4)[1], col="red")+geom_abline(slope=coef(Baby.lm4)[2], intercept=coef(Baby.lm4)[1]+coef(Baby.lm4)[3], col="blue")

Graph0+geom_smooth(method="lm", aes(group=Gender))
`geom_smooth()` using formula 'y ~ x'

Make the tables of fitted values

NewBabies  |> kable(caption="Fitted values from five models fitted to the Baby Weights data.", col.names=c("Age", "Gender", paste("Model",1:5)) )
Fitted values from five models fitted to the Baby Weights data.
Age Gender Model 1 Model 2 Model 3 Model 4 Model 5
36 Female 2967.667 2911.333 2674.032 2578.874 2552.733
38 Female 2967.667 2911.333 2905.089 2820.663 2813.533
40 Female 2967.667 2911.333 3136.145 3062.451 3074.333
42 Female 2967.667 2911.333 3367.202 3304.240 3335.133
36 Male 2967.667 3024.000 2674.032 2741.913 2762.707
38 Male 2967.667 3024.000 2905.089 2983.702 2986.672
40 Male 2967.667 3024.000 3136.145 3225.491 3210.638
42 Male 2967.667 3024.000 3367.202 3467.279 3434.603