|
|
Titanic: Statistical Analytics from Disaster |
Joao Negreiros, Samia Loucif, Mohammad Amin Kuhail, Ahmed Seffah
|
Abstract:
This study aims at utilizing statistical models to analyze the sort of passengers who likely survived the Titanic shipwreck in April 1912. A comparison between three approaches (logistic regression, decision tree, neural networks) was accomplished, including an exploratory analysis of the given dataset (composed of nine variables). Since the performance of those methods was quite similar, path analysis was computed to set up a statistical model for the survival (or not) factors of the RMS Titanic event. According to our calculations, this approach only explains 51.7% of the survival aftermath. Positively, other qualitative and quantitative critical factors were not included here.
|
|
| | |