lr: Logistic Regression

In jakespiteri/chicagocrime: Organises and Explores the Chicago Crime Dataset

Description

lr is used to fit a logistic regression model for a binary response variable.

Usage

lr(formula, data, init = NULL)

Arguments

formula	an object of class formula, a symbolic description of the model to be fitted. The specified names need to also be in data.
data	a required data frame containing the variables in the model
init	optional initial conditions to be passed to optimisation of the log-likelihood

Details

The form of the formula argument will be of the form response ~ predictor1 + predictor2 + ..., with predictor1 and predictor2 being named columns of the data frame in data.

The log-likelihood (from loglik_lr) is maximised using irls.lr with initial estimates given by init. If no initial values are supplied, this uses a vector of zeros instead.

Value

An S3 object of class 'lr', which is a list containing

coefficients	a vector of coefficients corresponding to covariates specified in formula
data	the data input to the function
formula	the formula input to the function
X	the model matrix X
val	value of the final log-likelihood at the values of coefficients, given by loglik_lr
its	number of iterations performed to retrieve the maximised log-likelihood

Examples

y = sample(0:1, 50, replace=TRUE)
d = data.frame(y = y, x = rnorm(10*y + 15))
fit = lr(y ~ x, data = d)

jakespiteri/chicagocrime documentation built on Jan. 24, 2020, 1:26 a.m.