We considered the variables smoke and parity, one at a time, in modeling birth weights of babies in Exercises 8.1 and 8.2. A more realistic approach to modeling infant weights is to consider all possibly related variables at once. Other variables of interest include length of pregnancy in days (gestation), mother’s age in years (age), mother’s height in inches (height), and mother’s pregnancy weight in pounds (weight). Below are three observations from this data set.
The summary table below shows the results of a regression model for predicting the average birth weight of babies based on all of the variables included in the data set.
We are using multivariable regression to help us evaluate the relationship between a predictor variable and the outcome while controlling for the potential influence of other variables.
\(\widehat{\mathrm{Birth Weight}} = \beta_{0} + \beta_{1}*gestation + \beta_{2}*parity + \beta_{3}*age + \beta_{4}*height + \beta_{5}*weight + \beta_{6}*smoke\)
\(\widehat{\mathrm{Birth Weight}} = -80.40 + \beta_{1}*.44 + \beta_{2}*-3.33 + \beta_{3}*-.01 + \beta_{4}*1.15 + \beta_{5}*.05 + \beta_{6}*-8.4\)
Each day of gestation increases baby birth weight by .44 ounces.
Each additional year on the mother’s age decrerases baby birth weight by .01 ounces.
Parity is a binary variable where it is 0 if the child is the first born, and 1 otherwise. On its own, the relationship demonstrates that children that are not first born are 1.93 lbs lighter than the first borns. The intercept is 120.07 ounces. However, our multivariable model has a intercept of -80 and includes other variables that shift the relationship so that parity is not our only variable.
predicted_1 <- -80.40 + (284*.44) + (0*-3.33) + (27*-.01) + (62*1.15) + (100*.05) + (0*-8.4)
predicted_1
## [1] 120.59
actual_1 <- 120
residual <- actual_1 - predicted_1
residual
## [1] -0.59
We use an adjusted \(R^2\) when there are many variables
r_var <- 249.28
out_var <- 332.57
n <- 1236
k <- 6
r2 <- 1 - (r_var/out_var)
r2
## [1] 0.2504435
r2adj <- 1 - ((r_var/out_var) * ((n-1)/(n-k-1)))
r2adj
## [1] 0.2467842