UCLA Accurate Prediction of A Low Weight Newborn Statistic Machine Worksheet
Question Description
Here are 3 question for statistic machine learning class.
(Use R for data analysis)
Data set birthwt in R package MASS was collected at Baystate Medical Center inSpringfield, MA. It contains the following variables (among others).
low: indicator of birth weight less than 2.5 kg
age: mothers age in years
lwt: mothers weight in pounds before pregnancy
smoke: smoking status during pregnancy (1 = smokes, 0 = does not smoke)ptl: number of previous premature labours
ht: history of hypertension (1 = present, 0 = absent)
(a) A 25-year old pregnant woman weighed 130 pounds before pregnancy. She smokes andhas no history of hypertension or premature labours. Use the given data and your favoritemethod to predict her probability of delivering a low-weight baby.
(b) Use your favorite cross-validation method to compare the prediction classification ratesof
logistic regression,
linear discriminant analysis,
quadratic discriminant analysis.
Which of these methods gives the most accurate prediction of a low-weight newborn?You may start your R code with commands
> install.packages(MASS); > library(MASS); > attach(birthwt);
Have a similar assignment? "Place an order for your assignment and have exceptional work written by our team of experts, guaranteeing you A results."