Day 25 - 10/18/2024

From last class

• Review on model design and model selection.
  • 1-What is the objective of the study?
    
  • 2-What is the data generating process?
    
  • 3-Think of a set of assumptions that represent (2) and can help answer (1). Build a statistical model.
    
  • 4-Fit the model to the data.
    
  • 5-Check that the assumptions were not that off.
    
• Always from simple to more complex!

Let’s talk about the semester projects.

• What is expected.
  
• What is not expected.

An example:

• Background: We want to investigate how different legume species perform in flooded conditions. We grew 5 different species under greenhouse conditions and divided them into 2 groups: treated and control. The treated were flooded for 39 days. The plant’s biomass was recorded at 3 points in time: at the time of application of the treatment, the day the treatment was concluded, and after a recovery of 43 days. The plants would be used as forage at this last point in time, i.e., day 237.
  
• Objective: We wish to find out which species yielded more biomass, and evaluate the tolerance to flooded conditions.
  
• Write out the statistical model.
  
• Which parameters are we interested in estimating?

Let’s try it out! R code. tutorial html

For next class

• We’ll focus on mixed models.
  
• A comment about the survey.