Inference using Stan

## Goal: make first mockup, communicate with pysd dev. team (due 7/31)

## Done
- Drafting information map between SD models with Hierarchical Bayesian models: table [here](https://docs.google.com/spreadsheets/d/1gbtGAA2YUaxXgHegiyoaa0WVKfPxnIgd7pxvXZABcIA/edit#gid=1633439295)

- Drafting computational map between`.xmile(mdl)` to `.stan`: detailed progress in Shinyoung's [pysd pj board](https://github.com/Dashadower/pysd/projects/2)
- classified SD variables into endogenous and exogenous and defined seven datatypes (endo: observed and simulated outcome (level var), simulated latent (aux. var) / exo: assumed parameter, estimated parameter, parapredictor, )parameters into time, rate, ratio, function

## Doing
- Jair's review on info.map and Angie's review on comp.map
- Shinyoung and Angie run Stan's `.sample` with externally generated predictor data
- Jair and Angie document Inventory model like that of prey-pradator and SIR model below:
-- SIR-family model:[stancode for SIR](https://github.com/stan-dev/example-models/blob/71e3dee590632e615b17b935652c68b69df6df6f/knitr/disease_transmission/boarding_school_case_study.Rmd#L356) (in R)
-- Lotka-Volterra (LV) prey-predator model: [notebook](https://github.com/StanKorea/casestudy/blob/master/lotka-volterra/20Stan_lotka_volterra.ipynb), [inference_stan](https://github.com/StanKorea/casestudy/blob/master/lotka-volterra/lotka-volterra.stan), [sbc_stan](https://github.com/StanKorea/casestudy/blob/master/lotka-volterra/sbc_lotka-volterra.stan) (in Python)

## TODO
- refine lookup function (possible connection with nonlinear fitting python scripts - more detail in pysd pj board)