Predicting fat percentage in the human body

Course Project. Developing Data Products. Johns Hopkins University.
Coursera Data Science Specialization.

• This application helps people to have an estimate of their fat percentage.
• This is a public application, that you can find in here: https://chechir.shinyapps.io/PredictFat/
• You need to input some data to the application
  • Age (yrs)
  • Weight (lbs)
  • Height (inches)
  • And other messurments such as your neck and abdomen circunference (in cms)

• The application provides two different estimations
  • Brozek's equation: BF = (4.57/ρ − 4.142) × 100
  • Siri's equation: BF = (4.95/ρ − 4.50) × 100
  • More details in here: https://en.wikipedia.org/wiki/Body_fat_percentage

• The application was programmed in R
• It uses the fat data that comes in the UsingR package
• It uses a multivariate linear regression
• Because we want to make the life easier for the users, the models only were trained with some columns
• Model R squared and adjusted R squared:

0.7447057 /  0.7363009

• As mentioned, this data is provided via the UsingR package in R
• It contains many physical measurements of 252 males.
• Most of the variables can be measured with a scale or tape measure.
• Can they be used to predict the percentage of body fat? If so, this offers an easy alternative to an underwater weighing technique.

“The data are as received from Dr. Fisher. Note, however, that there are a few errors. The body densities for cases 48, 76, and 96, for instance, each seem to have one digit in error as can be seen from the two body fat percentage values…”