Course summary

Survival analysis is a collection of statistical and forecasting methods commonly used to predict the survival time and to estimate the probability of an event within a specific period of time. Its techniques are often applied in healthcare (e.g. to estimate the time of hospital readmission/discharge, probability of death during the specific time of the follow-up period etc.), bioinformatics (e.g. cancer survival based on gene expression data), business (e.g. customer churn prediction, customer lifetime value etc.), higher education (e.g. student retention forecasts), and manufacturing (e.g. predicting the time of failure for specific devices or components).

This 2-day instructor-led live training course has been designed to provide you with a deep understanding of statistical models (i.e. non-parametric, semi-parametric and parametric) as well as machine learning-based survival analysis methods and their implementations in the R programming language. The 1st day of the course comprehensively covers industry-standard survival analysis approaches such as Kaplan-Meier, Cox regression and more complex techniques e.g. time-dependent Cox models and regularised regressions. The 2nd day of the course is dedicated to machine learning and AI time-to-event methods applicable to high-dimensional censored data e.g. survival trees, Bayesian methods and more black-box approaches such as Support Vector Machines (SVMs) and tree-based ensembles (i.e. supermodels).

The presented tutorials will utilise datasets from a variety of fields: social sciences, biomedical sciences, economics and business. The course will implement custom-made R code as well as methods and functions available in selected R packages e.g. survival, BMA, fastcox, randomForestSRC, mboost, AER and many others.

Who is this course for?

This course will be especially useful to data scientists, business/risk analysts, econometricians, bioinformaticians as well as ambitious MSc students, PhD-level and post-doc researchers interested in learning and applying an array of industry-standard and cutting-edge survival analysis methods including machine learning and AI approaches to model and predict a number of problems based on censored data e.g. time to event, health/diagnostic outcomes e.g. survival forecasts, customer churn/retention, academic dropout rates, time-to-failure and similar.

Programme outline

During this course, you will:

• Understand the mathematical underpinnings and statistical assumptions of the hazard function typically used in survival analysis and computational complexity of the methods presented during the course,
• Estimate the hazard function with the Kaplan-Meier and Nelson-Aalen non-parametric methods, the semi-parametric Cox regression approach and parametric censored linear regression (e.g. Tobit or Buckley-James regressions) as well as the accelerated failure time (AFT) models,
• Implement more complex statistical survival methods e.g. time-dependent (i.e. with time-dependent covariates) and penalised Cox regression models including Lasso, Ridge and Elastic Net regularisation techniques,
• Apply cutting-edge machine learning and AI algorithms tailored to handle censored data e.g. survival trees with different spitting criteria, Bayesian methods (e.g. Naive Bayes and Bayesian networks), and the variants of Support Vector Machines (SVMs) and tree-based ensemble methods (e.g. bagging survival trees and random survival forests) specifically designed for survival analysis purposes,
• Evaluate the survival prediction performance with metrics designed for censored data e.g. the concordance index (C-index), mean absolute error (MAE), and Brier score,
• Visualise survival analysis results using R language e.g. with the Kaplan-Meier curve and other approaches.

What is included in the course fee?

At Mind Project, we pride ourselves for delivering great quality, goal-directed, interactive and cost-effective training. Our remote, hybrid, and on-premises tutor-led courses are designed to maximise your opportunity to learn complex topics in shortest possible time while being able to discuss and interact with our tutor and other fellow attendees. Therefore, the registration fee for this course includes a number of benefits that many other training providers don’t offer:

• 2 days of remotely delivered live tutorials, lectures and presentations with our expert tutor – you will be able to ask questions, interact and discuss the topic with other attendees from the comfort of your home/office,
• 1-year access to the course area on Mind Project Learning Platform with the recorded videos of the training sessions and additional course resources (i.e. research articles, book excerpts, and additional online resources),
• R code scripts, datasets and copy of presented materials used during the course,
• 2-week post-course follow-up period during which you will be able to work on homework exercises and submit them for our feedback – the feedback will be provided either via email or on Teams (i.e. as a 15-minute 1-2-1 session with our tutor),
• support throughout the course and the 2-week post-course follow-up period – feel free to ask us questions about the material presented at the training during this period.