The term "survival analysis"
1. Survival Analysis / Time-to-Event Analysis in R
Hi, my name Heidi Seibold, and I will be your instructor for this course. I am a statistician at the LMU in Munich. In this course, you will learn basic methods to analyze time-to-event data. We will keep the mathematics at a minimum and focus on HOW to do the analyses in R. After finishing this course you will be able to compute survival curves and common statistical models for continuous duration times. All methods we discuss are able to deal with right censoring (if you don't know what that means, no worries, we will discuss this soon!). You will also be able to visualize your results. We will not only focus on survival times, where the event of interest is death but also discuss other events, such as the time until someone finds a job again after becoming unemployed. Fortunately, the methods are the same.2. The term survival analysis
Before digging into the statistical methods that can be used for survival analyses, let's discuss what survival analysis actually means. Instead of calling this course "Survival analysis in R" we should actually have called it "Time-to-event data analysis in R" because we will discuss methods where the event does not have to be death. The reason why we still called it "Survival analysis in R" is because this is the most commonly used term. So what will we actually discuss in the course? We will discuss duration times. Usually, we are interested in a certain event (for example, death) and want to know the time until this event happens. In the graph here "x" shows the time point of the event for 5 individuals. Often times this event is death and the line visualizes the time-until-death from the start of a study. But let's go through a couple of other examples!3. The term survival analysis
In one of the data sets we will be working with, the starting point 0 is the time a person becomes unemployed. The event of interest is re-employment. The line shows the time needed for the individual to find a job again.4. The term survival analysis
The event could also be the delivery of a letter and we are interested in the time needed for a letter to be delivered.5. The term survival analysis
Or how long does it take for a cab to pick you up at your house after having called the cab company? As you see, the event of interest can be very diverse, but we are always interested in the time until this event occurs.6. Data sets we will be using
Two of the datasets we will be using in this course are the GBSG2 data set and the UnempDur data set. In the GBSG2 dataset, we are interested in the time until death in breast cancer patients. We have data of 686 patients including information on the treatment, tumor size, age, and other potentially important factors. The data set is available in the TH.data package and can be loaded with the data function. The UnempDur dataset contains information on individuals who have lost their jobs. We are interested in investigating how long it takes for them to find a full-time position again. It is available in the Ecdat package. In the upcoming exercises, we will take a deeper look into these datasets. For this, I already have a pro tip for you: To see information on the dataset, exchange the data() function with the help() function.7. Let's practice!
Time to look into the two datasets.Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.