37 Section Summary

Arguably this is the only section only the miniscule portion, in light of the whole mathematical models, was included.  It was the choice the author had to make given the limited space in this summary.  We only covered some aspects of empirical models and the gist of stochastic modeling with the use of Markov chain.

What has not been included is indeed generates a long list:

• Dynamic models involving differential equations were not discussed.  Often the interest is a time-dependent change, and one area this technique is widely utilized is in the modeling of communicable/infectious disease transmission.
• Linear programming (LP) or optimization was not discussed.  Linear programming also deals with systems of linear equations, but inequalities as well.  By utilizing methods such as introducing slack variables to transform the inequalities into equations, then LP essentially becomes linear algebra.  However, LP is different in that its objective is to find an optimal solution, not just a feasible solution.
• Simulation modeling was not discussed.  Random number generation, queuing theory, and Monte Carlo simulation are notable examples.
• The famous game theory, a branch of strategic modeling, was not discussed.  One area where game theory is actively studied is political science: the Nobel laureate economist and political scientist Kenneth Arrow proved in his Arrow’s impossibility theorem or general impossibility theorem, that there cannot exist a “fair” rank-order voting system.

The author in fact has experience in publishing scientific research papers, specifically in the field of medical science.  As master’s concentration, the author chose epidemiology over health policy, while the author’s interest better aligned with policy, the author feared it would be less scientific than more rigorous statistics-based epidemiology as seen in the curriculum.

However, it was the author’s critical misunderstanding, and the author ended up writing two health policy research papers.  What the author has learnt through research experience is every quantitative scientific research is the same in that it involves statistical testing, while there might be more preferred or less commonly used statistical methods.  No matter how subjective the nature of the content is, politics, psychology, or whatsoever, we have to, in some way, quantify the variables of interest and test the validity of our hypotheses.  That is, examine what we have observed is just a happenstance vs a logical necessity.

That was the reason why the author chose to discuss at least briefly a handful of modeling techniques instead of giving a definition or a very cursory overview of numerous methods.

Again, the rest is up to the reader, filling in the contents of each section the author introduced in this summary.