NMST531 - Censored Data Analysis (Exercise class)
SIS page : NMST531
Schedule : on Monday 12:20 - 13:50 in classroom K4
Language : materials in English, instructed in Czech (unless anybody requires English)
Instructor e-mail : vavraj@karlin.mff.cuni.cz (for sending solutions to assignments)
Lecture web page (from previous year): www.karlin.mff.cuni.cz/~kulich/vyuka/cens/index.html
Note : exercises will start by the end of October
CORONAVIRUS : We have to be ready for any potential government restrictions.
-
Presence form:
- weekly exercise class on schedule,
- completing given homework assignments in pre-specified deadlines (usually a day before the next exercise),
- exercise class will be opened with discussion over previous homework,
- we will go through the materials for exercise class together,
- there will be time for you to ask questions, work on given assigments, ...
- no attendance requirement for course credit.
-
Distant form:
- regular self-study of given materials,
- completing given homework assignments in pre-specified deadlines,
- feedback and discussions over the last exercise and homework via ZOOM session shortly after the deadline,
- also in ZOOM: assignment of a new homework for the next week with a short commentary,
- not in ZOOM: full lecture and detailed exploration of given materials (everybody has his/her own pace while working with R).
Credit
- Satisfactory solution to given assignments by the prescribed deadline.
(If your solution will not be satisfactory enough, you might be asked for revision.)
Hand in printed (or hand written) report/solution in the next exercise class or send via email to address vavraj@karlin.mff.cuni.cz.
Recommended form: Rmarkdown output (PDF or HTML)
Overview
Exercise 1 (7-14th November)
- Topic: Parametric regression models for censored data
- Dataset:
pbc
dataset fromlibrary(survival)
- Assignment: Find suitable model for survival time and interpret it.
- Deadline: Monday 14th November 9:00
Exercise 2 (14-21st November)
- Topic: Non-parametric estimation of cumulative hazard and survival function
- Dataset: fans
- Assignment: Calculate and plot [NA], [KM], [FH]
estimates both manually and using
survfit
function fromlibrary(survival)
. - Deadline: Monday 21st November 9:00
Exercise 3 (21-28th November)
- Topic: Actuarial (lifetable) estimator of the survival
- Dataset: nurshome,
mort.m
andmort.f
from mort.RData - Assignment: Compare non-grouped and grouped
approaches for estimation survival and hazard function for two
subpopulations in
nurshome
dataset. Compare survival estimators with the output from ČSÚ: males and females. - Deadline: Monday 28th November 9:00
Exercise 4 (28th November - 5th December)
- Topic: Confidence intervals and bands for survival function
- Dataset: km_all.RData
- Assignment: Compute and plot [KM] estimates including Hall-Werner bands. Perform short (or long if you wish) simulation study.
- Deadline: Monday 5th December 9:00
Exercise 5 (5-13th December)
- Topic: Testing equality of censored distributions
- Datasets: km_all.RData
- Assignment: Calculate and plot [KM], [NA] estimates and smoothed estimator of hazard function when differentiating different groups. Perform two-sample tests and decide (based on the plots) which test statistic would be the most appropriate.
- Deadline: Tuesday 13th December 9:00 (also for Ex 6!)
Exercise 6 (5-13th December)
- HELD TOGETHER WITH Ex5 on 5th December at 12:20!
- Topic: The choice of two-sample test statistic
- Datasets:
data(veteran)
, nurshome - Assignment: Fill table of appropriateness of different weights in two-sample survival tests in different situations. Perform several real data analyses and simulation study.
- Deadline: Tuesday 13th December 9:00 (also for Ex 5!)
Exercise 7 (13th December - 2nd January)
- Held on Tuesday 13th December at 14:00!!! On Monday 12th December there will be a lecture on Cox model by doc. Hlubinka.
- Topic: Building Cox models for censored data (constant covariates)
- Dataset:
pbc
dataset fromlibrary(survival)
- Assignment: Perform exploratory analysis focused on the influence of covariates on the survival probability. Build a reasonable Cox model (starting from simple univariate models). Compare your final Cox model to your final model from Exercise 1.
- Deadline: Monday 2nd January 9:00
Bonus Exercise (not scheduled)
- Topic: Generalizations of the Cox model
- Dataset:
pbc
andcgd
datasets fromlibrary(survival)
- Assignment: None, just admire what else could be done with the Cox model.
Exercise 8 (2-9th January)
- Topic: Time-varying covariates in the Cox model
- Dataset:
jasa
andheart
dataset fromlibrary(survival)
- Assignment: Try to reproduce
heart
dataset fromjasa
. Build and interpret Cox model with covariate indicating the time of heart transplantation and other fixed covariates. Does transplantation help patients to survive longer? Plot estimated survival functions. - Deadline: Monday 16th January 9:00