Die Bausteine für Bayes'sche Inferenz

Grundlagen der Bayes'schen Datenanalyse in R

Rasmus Bååth

Data Scientist

Was ist ein generatives Modell?

Generatives Zombie-Drogen-Modell

# Parameters
prop_success <- ???
n_zombies <- ???

Generatives Zombie-Drogen-Modell

# Parameters
prop_success <- 0.15
n_zombies <- 13

Generatives Zombie-Drogen-Modell

# Parameters
prop_success <- 0.15
n_zombies <- 13
# Simulating data
data <- c()
for(zombie in 1:n_zombies) {
  data[zombie] <- ???
}

Generatives Zombie-Drogen-Modell

# Parameters
prop_success <- 0.15
n_zombies <- 13
# Simulating data
data <- c()
for(zombie in 1:n_zombies) {
  data[zombie] <- runif(1, min = 0, max = 1) < prop_success
}

Generatives Zombie-Drogen-Modell

# Parameters
prop_success <- 0.15
n_zombies <- 13
# Simulating data
data <- c()
for(zombie in 1:n_zombies) {
  data[zombie] <- runif(1, min = 0, max = 1) < prop_success}

data

FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE

Generatives Zombie-Drogen-Modell

# Parameters
prop_success <- 0.15
n_zombies <- 13
# Simulating data
data <- c()
for(zombie in 1:n_zombies) {
  data[zombie] <- runif(1, min = 0, max = 1) < prop_success
}
data <- as.numeric(data)

Generatives Zombie-Drogen-Modell

# Parameters
prop_success <- 0.15
n_zombies <- 13
# Simulating data
data <- c()
for(zombie in 1:n_zombies) {
  data[zombie] <- runif(1, min = 0, max = 1) < prop_success
}
data <- as.numeric(data)
data

0 0 0 1 0 0 0 0 1 0 1 0 0

Generatives Zombie-Drogen-Modell

# Parameters
prop_success <- 0.15
n_zombies <- 13
# Simulating data
data <- c()
for(zombie in 1:n_zombies) {
  data[zombie] <- runif(1, min = 0, max = 1) < prop_success
}
data <- as.numeric(data)
data

0 0 1 0 0 0 0 0 0 0 0 0 0

Generatives Zombie-Drogen-Modell

# Parameters
prop_success <- 0.15
n_zombies <- 13
# Simulating data
data <- c()
for(zombie in 1:n_zombies) {
  data[zombie] <- runif(1, min = 0, max = 1) < prop_success
}
data <- as.numeric(data)
data

0 1 0 1 1 0 0 1 0 1 0 0 0

Generatives Zombie-Drogen-Modell

# Parameters
prop_success <- 0.15
n_zombies <- 13
# Simulating data
data <- c()
for(zombie in 1:n_zombies) {
  data[zombie] <- runif(1, min = 0, max = 1) < prop_success
}
data <- as.numeric(data)
data

0 0 0 0 0 0 0 1 0 0 0 0 0

Generatives Zombie-Drogen-Modell

# Parameters
prop_success <- 0.15
n_zombies <- 13
# Simulating data
data <- c()
for(zombie in 1:n_zombies) {
  data[zombie] <- runif(1, min = 0, max = 1) < prop_success
}
data <- as.numeric(data)
data

0 0 0 0 1 0 0 0 0 1 0 1 0

Probier dieses Modell aus!

Grundlagen der Bayes'schen Datenanalyse in R