Joshua Entrop - Blog

The 3 rules of do-calculus

causal inference

Pearl’s do-calculus offers a comprehensive set of rules for identifying causal effects from a causal directed acyclic graph (DAG). Using those rules, one can identify causal effects even in situations in which the commonly applied back-door criteria does not hold. In this blog post, I demonstrate how you can use Pearl’s do-calculus to identify causal effects based on various example DAGs.

Joshua Philipp Entrop

Testing degrees of freedom for generalised survival models in R

R

survival analysis

Have you ever encountered a situation in which you wanted to test different degrees of freedom (DF) for a generalised survival model? This blog post introduces an R function that helps use finding the best fitting spline functions for you generalised survival model.

Joshua Philipp Entrop

{ExclusionTable} a package for keeping track of exclusions and inclusions

R

ExclusionTable

In today’s blog post we will take a look at a package that allows you to keep track of the number of observations that you in- or exclude from your dataset or study population. For this I will introduce the ExclusionTable package, which provides a usufull function this job.

Joshua Philipp Entrop

Optimisation of a stratified Cox model using Optimx()

Optimisation

R

survival analysis

In this blog post we are going to fit a stratified Cox regression model by optimising its likelihood function with Optimx::optimx(). Stratified Cox regression models allow one to relax the assumption of proportional hazards over time between different exposure groups.

Joshua Philipp Entrop

Optimisation of a Cox proportional hazard model using Optimx()

Optimisation

R

survival analysis

This blog post explains how one can manually optimise a Cox proportional hazard model using a maximum likelihood estimation (MLE) method in R without the use pre-defined functions.

Joshua Philipp Entrop

Comparing the confidence intervals of a Weibull model estimated with flexsurvreg() and optimx()

Optimisation

R

survival analysis

This blog post is a follow up on my previous post on optimising a Weibull regression model using optimx(). This time I’ll try to find a solution for the discrepancy between the confidence interval estimates of the Weibull hazard function estimated with optimx() and flexsurvreg().

Joshua Philipp Entrop

Optimisation of a Weibull survival model using Optimx() in R

Optimisation

R

survival analysis

In this blog post we will optimise a Weibull regression model by maximising its likelihood function using optimx() from the {optimx} package in R.

Joshua Philipp Entrop

Optimisation of a Poisson survival model using Optimx in R

Optimisation

R

In this blog post, we will fit a Poisson regression model by maximising its likelihood function using optimx() in R. As an example we will use the lung cancer data set included in the {survival} package.

Joshua Philipp Entrop

Estimating Standard Errors for a Logistic Regression Model optimised with Optimx in R

Optimisation

R

In my previous I estimated the point estimates for a logistic regression model using optimx() from the optimx package in R. In this post I would like to contine with this model an try to find the standard error (SE) for the derived estimates.

Joshua Philipp Entrop

Optimisation of a Logistic Regression Model using Optimx in R

Optimisation

R

This blog post explains how one can manually optimise a logistic regression model using a maximum likelihood estimation (MLE) method in R without the use pre-defined functions.

Joshua Philipp Entrop

Optimisation of a Linear Regression Model in R

Optimisation

R

In this post I would like to show how to manually optimise a linear regression model using the optim() command in R. Usually if you learn how to fit a linear regression model in R, you would learn how to use the lm() command to do this. However, if you would like to know how to do this manually, examples are rare.

Joshua Philipp Entrop