Investigate adding simple baselines

[mgdenno]

There is a need to add simple baselines "models" to TEEHR such as climatology, persistence, daily mean, etc. There are also some that are more complex, but still "simple". How to implement this?

• Option 1: My gut says this is the way to go. We add some methods to generate timeseries and store them as configurations such that they can be "evaluated" as a baseline.
  • I think for persistence we could create a eval.generate_persistence() method that would generate a persistence forecast and add it to the database with the configuration_name = persistence. What arguments would this need? Is it only valid/useful when there are actual forecasts (i.e. reference_time is not NULL), or is it also useful for simulations (I'd think no, but, maybe?)
  • Daily mean. I think again we could add a `eval.generate_daily_mean()' method. Not sure exactly what this represents (historic daily mean for that day of the year, for example)? If that is the case, that sounds like a reference timeseries, or if the period of record is sufficiently long we may be able to calculate it internally. Do we need/want both options?
  • Climatology- Is there a specific calculation for this or is it a reference timeseries that is needed?
  • What others can/should we add?
• Option 2: make it happen on the fly. Not totally sure how this would work, but maybe when calculating metrics, we could generate this type of result. Right now I favor Option 1.