I am interested in the use of modern causal inference methods to research the association between a (non-genetic) exposure, and endogenous molecules and/or health outcomes (high-dimensional data). I came upon the work of Prof. Mark J. van der Laan and his group in the USA, and it seems that it's probably what I need (but correct me if I am wrong). I decided to go through his book, Targeted Learning in Data Science, but I am facing multiple difficulties. Namely, I am not really used to the notation and some of the concepts are quite unclear to me. Can you suggest me any reference I can use as background material before tackling it? My background is on bioinformatics and I am currently doing PhD-level research in Epidemiology: I took courses in basic statistics and machine learning and I am reading the What If book.
1$\begingroup$ See my answer here: stats.stackexchange.com/questions/568281/…. Then the authors' own Targeted Learning book appears to precede (possibly even be a prerequisite for) Targeted Learning in Data Science. $\endgroup$– Adrian KeisterMar 24, 2022 at 16:48
1$\begingroup$ These notes by Edward Kennedy go over much of the relevant background material. arxiv.org/pdf/2203.06469 arxiv.org/pdf/1510.04740.pdf $\endgroup$– LarsApr 15, 2022 at 2:44