heart <- [Link]("heart.

dat", quote = "\"")

names(heart) <- c("AGE", "SEX", "CHESTPAIN", "RESTBP", "CHOL",
"SUGAR", "ECG", "MAXHR", "ANGINA", "DEP", "EXERCISE", "FLUOR",
"THAL", "OUTPUT")

# ---
heart$CHESTPAIN = factor(heart$CHESTPAIN)
heart$ECG = factor(heart$ECG)
heart$THAL = factor(heart$THAL)
heart$EXERCISE = factor(heart$EXERCISE)
heart$OUTPUT = heart$OUTPUT - 1

# ---
library(caret)
[Link](987954)
heart_sampling_vector <-
createDataPartition(heart$OUTPUT, p = 0.85, list = FALSE)
heart_train <- heart[heart_sampling_vector,]
heart_train_labels <- heart$OUTPUT[heart_sampling_vector]
heart_test <- heart[-heart_sampling_vector,]
heart_test_labels <- heart$OUTPUT[-heart_sampling_vector]

heart_model <-
glm(OUTPUT ~ ., data = heart_train, family = binomial("logit"))
summary(heart_model)

pnorm(3.362 , [Link] = F) * 2

heart_model2 <- glm(OUTPUT ~ AGE, data = heart_train, family =

binomial("logit"))
summary(heart_model2)

log_likelihoods <- function(y_labels, y_probs) {

y_a <- [Link](y_labels)
y_p <- [Link](y_probs)
y_a * log(y_p) + (1 - y_a) * log(1 - y_p)
}

dataset_log_likelihood <- function(y_labels, y_probs) {

sum(log_likelihoods(y_labels, y_probs))
}

# ---

deviances <- function(y_labels, y_probs) {

-2 * log_likelihoods(y_labels, y_probs)
}

dataset_deviance <- function(y_labels, y_probs) {

sum(deviances(y_labels, y_probs))
}

type parameter:
model_deviance <- function(model, data, output_column) {
y_labels = data[[output_column]]
y_probs = predict(model, newdata = data, type = "response")
dataset_deviance(y_labels, y_probs)
}
model_deviance(heart_model, data = heart_train, output_column =
"OUTPUT")
null_deviance <- function(data, output_column) {
y_labels <- data[[output_column]]
y_probs <- mean(data[[output_column]])
dataset_deviance(y_labels, y_probs)
}
null_deviance(data = heart_training, output_column = "OUTPUT")

model_pseudo_r_squared <- function(model, data, output_column) {

1 - ( model_deviance(model, data, output_column) /
null_deviance(data, output_column) )
}

model_pseudo_r_squared(heart_model, data = heart_train,

output_column = "OUTPUT")

model_chi_squared_p_value <- function(model,

data, output_column) {
null_df <- nrow(data) - 1
model_df <- nrow(data) - length(model$coefficients)
difference_df <- null_df - model_df
null_deviance <- null_deviance(data, output_column)
m_deviance <- model_deviance(model, data, output_column)
difference_deviance <- null_deviance - m_deviance
pchisq(difference_deviance, difference_df,[Link] = F)
}

model_chi_squared_p_value(heart_model, data = heart_train,

output_column = "OUTPUT")
model_deviance_residuals <- function(model, data, output_column) {
y_labels = data[[output_column]]
y_probs = predict(model, newdata = data, type = "response")
residual_sign = sign(y_labels - y_probs)
residuals = sqrt(deviances(y_labels, y_probs))
residual_sign * residuals
}
summary(model_deviance_residuals(heart_model, data =
heart_train, output_column = "OUTPUT"))
train_predictions <- predict(heart_model, newdata = heart_train,
type = "response")
train_class_predictions <- [Link](train_predictions 0.5)
mean(train_class_predictions == heart_train$OUTPUT)

# ---
test_predictions = predict(heart_model, newdata = heart_test,
type = "response")
test_class_predictions = [Link](test_predictions 0.5)
mean(test_class_predictions == heart_test$OUTPUT)

library(glmnet)
heart_train_mat <- [Link](OUTPUT ~ ., heart_train)[,-1]
lambdas <- 10 ^ seq(8, -4, length = 250)
heart_models_lasso <- glmnet(heart_train_mat,
heart_train$OUTPUT, alpha = 1, lambda = lambdas, family = "binomial")
[Link] <- [Link](heart_train_mat, heart_train$OUTPUT,
 alpha = 1,lambda = lambdas, family = "binomial")
lambda_lasso <- [Link]$[Link]
lambda_lasso

predict(heart_models_lasso, type = "coefficients", s = lamb-da_lasso)

lasso_train_predictions <- predict(heart_models_lasso,

s = lambda_lasso, newx = heart_train_mat, type = "response")
lasso_train_class_predictions <-
[Link](lasso_train_predictions 0.5)
mean(lasso_train_class_predictions == heart_train$OUTPUT)

# ---
heart_test_mat <- [Link](OUTPUT ~ ., heart_test)[,-1]
lasso_test_predictions <- predict(heart_models_lasso,
s = lambda_lasso, newx = heart_test_mat, type = "response")
lasso_test_class_predictions <-
[Link](lasso_test_predictions 0.5)
mean(lasso_test_class_predictions == heart_test$OUTPUT)

# ---

(confusion_matrix <- table(predicted =

train_class_predictions, actual = heart_train$OUTPUT))
actual
predicted 0 1
0 118 16
1 10 86
(precision <- confusion_matrix[2, 2] / sum(confusion_matrix[2,]))

# ---
(recall <- confusion_matrix[2, 2] / sum(confusion_matrix[,2]))

# ---
(f = 2 * precision * recall / (precision + recall))

# ---

(specificity <- confu-sion_matrix[1,1]/sum(confusion_matrix[1,]))

# ---

library(ROCR)
train_predictions <- predict(heart_model, newdata = heart_train,
type = "response")
pred <- prediction(train_predictions, heart_training$OUTPUT)
perf <- performance(pred, measure = "prec", [Link] = "rec")

thresholds <- [Link](cutoffs = perf@[Link][[1]], recall =

perf@[Link][[1]], precision = perf@[Link][[1]])
subset(thresholds,(recall 0.9) & (precision 0.8))

glass <- [Link]("[Link]", header = FALSE)

names(glass) <- c("id","RI","Na", "Mg", "Al", "Si", "K", "Ca",
"Ba", "Fe", "Type")
glass <- glass[,-1]
[Link](4365677)
glass_sampling_vector
 <- createDataPartition(glass$Type, p = 0.80, list = FALSE)
glass_train <- glass[glass_sampling_vector,]
glass_test <- glass[-glass_sampling_vector,]

library(nnet)
glass_model <- multinom(Type ~ ., data = glass_train, maxit = 1000)
summary(glass_model)

glass_predictions <- predict(glass_model, glass_train)

mean(glass_predictions == glass_train$Type)

# ---
table(predicted = glass_predictions, actual = glass_train$Type)

glass_test_predictions <- predict(glass_model, glass_test)

mean(glass_test_predictions == glass_test$Type)

# ---
table(predicted = glass_test_predictions, actual =
glass_test$Type)

wine <- [Link]("[Link]", sep = ";")

wine$quality <- factor(ifelse(wine$quality < 5, 0,
ifelse(wine$quality 6, 2, 1)))
[Link](7644)
wine_sampling_vector <- createDataPartition(wine$quality, p =
0.80, list = FALSE)
wine_train <- wine[wine_sampling_vector,]
wine_test <- wine[-wine_sampling_vector,]
library(MASS)
wine_model <- polr(quality ~ ., data = wine_train, Hess = T)
summary(wine_model)
[Link](table(wine$quality))

wine_predictions <- predict(wine_model, wine_train)

mean(wine_predictions == wine_train$quality)

# ---
table(predicted = wine_predictions,actual = wine_train$quality)

wine_test_predictions <- predict(wine_model, wine_test)

mean(wine_test_predictions == wine_test$quality)

# ---
table(predicted = wine_test_predictions,
actual = wine_test$quality)
wine_model2 <- multinom(quality ~ ., data = wine_train,
maxit = 1000)
wine_predictions2 <- predict(wine_model2, wine_test)
mean(wine_predictions2 == wine_test$quality)

# ---
table(predicted = wine_predictions2, actual = wine_test$quality)
actual
AIC(wine_model)
# ---
AIC(wine_model2)

# ---