Sent in the final game plan, started working on proposals.
Mar 5, 2019
Today we talked with the client again. Even if I think the client wanted to evaluate the causal effect of the fault detection warnings, both the leading professor on my team and the client wanted to just do association / prediction. From my understanding, even if association will not get us closer to the causal result, they still want to stick with it, since it feels comfortable to most people in the group. I might not be able to explain it well enough, they also have their strong mind not to be changed, so I plan to give up. If I do not write the statement of the result in the end, I just hope my name do not appear in the document, and we add a "we do not care" as a caveat to the project XD.
Today's class Professor Wasserman mentioned a book called "the Book of Why" by Judea Pearl. Sounds pretty interesting. The other day Professor Edson Severnini mentioned a book "Freakanomics", which might be helpful to explain to people how causation is different from prediction.
The game plan document is still slow in progress, but I hope to get it out quicker.
Mar 4, 2019
Today I read about energy savings performance contract (ESPC), while writing about my game plan. It occurs to me that the projects listed as "ESPC" in GSA data set are very likely to have a contract documenting the projected savings. If this is the case, I might be able to do a comparative study of quasi-experimental evaluation vs ex-ante projected savings, following Fowlie et al.