I am often asked if logistic regression is a machine learning algorithm. I say that it is not, for I can formulate it mathematically and solve it using matrix equations, for example. Its solution is derived deterministically and estimation is performed mathematically, through optimization methods. Its link function is a is a mathematical equation.
Logistic regression is most appropriate when the dependent variable (target variable) has two possible outcomes (binary). Will customers respond to an offer or unsubscribe, will the enemy fight or flee, will subjects respond to treatment or grow ill, will livestock live or die? Yes or no? One or zero?
Here I will take you on a journey into the art and science of predictive modeling using logistic regression, inside-and-out.
|Imprint||Glasstree Academic Publishing|
|Copyright||2017, Jeffrey Strickland|
|Copyright License||Standard Copyright License|
|Product Details||6 x 9 Standard Mono Matte Perfect Bound|
|Page Count||334 pages|
|Type of Publication||Textbook|
|Peer Review Status||Open, Completed|
|Keywords||logit, logistic regression, modeling, binary response, R Programming, R Studio, SAS Studio, Python, analytics, data analytics|