The data I'm working on consists of several components that are arguably correlated with each other. In particular, the study collects information of health care facilities related to the organizing and system of care for a disease, including service delivery, revenues and spending, human resources, policy, training, etc. So either there are studies examining relationships between similar components or conducted at other settings, or we have good reasons to hypothesize some correlation among those constructs/components.

What I want to do is to analyze these components as a system in the sense that they are inter-connected and that change in one sector not only is subject to effects of other sectors but also leads to changes in those sectors. I want to go beyond my "traditional" approach; that is, to select and examine relationship between a certain variable and a set of its predictors. I feel like doing so wouldn't be the best way to gain a good overview of the system, nor would that be sufficient in understanding the complex relationship nature of the system's constituting elements.

But I've never done such kind of "whole system" analysis, so I don't know which sources of reference to look at. I try to do some search and it looks like my analysis intention may cut across a few domains new to me such as simulation and system analysis (but again that's just my search and thought, it may be right or wrong).

So I'd welcome and appreciate any inputs and comments, e.g., what you think about analytical approaches for this sort of topic and where I may look at and gain knowledge in order to proceed with specifying the analysis plan.


You might be interested about simultaneous equation modeling or structural equation modeling which have a long tradition in econometrics dating at least from the 1940's.

There is a R language package "sem" which implements these methods. Look for example Klein data of US in that package and read some literature.

W. Greene's Econometric Analysis might be a good book for building basic understanding in this area.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.