Predict your child's adult height

Vedrana B.

The Data Science Specialization: Developing Data Products

PROJECT OVERVIEW

  • Problem description:

    • predict a child's adult height based on parent's heights

 

  • Problem solution:

    • using existing Galton's dataset (1885 study)

      • heights of 928 children and their 205 parents (in inches)

      • parents represented by midparent height:

        \(\qquad \qquad hParent = \frac{hFather + 1.08*hMother}{2}\)

    • linear model fit on the Galton's dataset:

      \(\qquad \qquad \qquad hChild = \alpha * hParent + \beta\)

    • SHINY application available online

GALTON Dataset preview

# Chunk of R code for plotting interactive rCharts scatterplot
library(UsingR); require(base64enc); require(rCharts)
data(galton)
options(RCHART_WIDTH = 600, RCHART_HEIGHT = 300)
knitr::opts_chunk$set(comment = NA, results = 'asis', tidy = F, message = T)

g1 <- nPlot(child ~ parent, data = galton, type = 'scatterChart')
#g1$show('inline', include_assets = TRUE)
g1$save('fig/g1.html')
cat('<iframe src="fig/g1.html" width=100%, height=600></iframe>')

Linear prediction model (LM)

  • Linear relationship between input variable (parent height) and the output (child height)
# Chunk of R code for building the LM model and for predicting:
model <- lm(formula = child ~ parent, data = galton2)
p     <- (as.numeric(input$hF) + 1.08*as.numeric(input$hM))/2
c     <- predict(model, data.frame(parent = p))

# Chunk of R code for ploting the linear fit to the Galton data (inches):
library(ggplot2)
limits <- c(min(galton)-1,max(galton)+1)
ggplot(data = galton, aes(x=parent,y=child)) + 
    geom_point(color = "red", alpha=0.2, size=3) + geom_smooth(method='lm')  +
    labs(x = "Parent\'s height", y = "Child\'s height", title ="LM prediction using Galton\'s dataset") + 
    coord_cartesian(xlim = limits, ylim = limits) + guides(color = FALSE, fill = FALSE)

plot of chunk plot2

Interactive SHINY web app

REFERENCES