ex5_2_6.R

# exercise 5.2.6

# Load wine data 
source(file.path("Scripts", "ex5_1_5.R"))
X = data.frame(X)
xnam <- paste("X", 1:dim(X)[2], sep="")
(fmla <- as.formula(paste("y ~ ", paste(xnam, collapse= "+"))))

# Fit logistic regression model to predict the type of wine
w_est = glm(fmla, family=binomial(link="logit"), data=X);


# Define a new data object with the attributes given in the text
x = data.frame(cbind(6.9, 1.09, .06, 2.1, .0061, 12, 31, .99, 3.5, .44, 12))
colnames(x) <- xnam

# Evaluate the logistic regression for the new data object
p = predict(w_est, newdata=x, type="response")