5 Day 5 (June 13)

5.1 Announcements

• Grades should be up-to-date

  • Remember that a paragraph is not 1 sentence.
  • I want the html or pdf not the Rmd for Assignment 1

• Donut/extra credit day!

• Assignment 2 is posted

• Recommended reading

  • Chapters 1 and 2 (pgs 1 - 27) in Linear Models with R

• Selected questions from journals

  • What exactly is a parameter?
  • What is \(\beta\) doing in the model?
  • “I’m getting confused about predictor variables - how do you put multiple predictor variables on a graph?”
  • “Why exactly do we want to estimate \(\beta\)”
  • Are you struggling with R programming?
  • What is a a matrix inverse?
  • What is matrix algebra used for?
  • “One thing I’m still confused by is telling the difference between a linear and a nonlinear model. I know you explained it in detail, it’s hard to wrap my head around. So, you can change up the predictor variable and still have it represent a linear model, but once you change the parameters, it becomes nonlinear? Am I getting this right?”
  • “I’m struggling to understand when and why I should choose a least squares approach versus a likelihood-based method when building models. Can I have other distributions than normal when using the minimize a loss function? Or, can I assume that when I have a normal, binomial, Poisson distribution I should use maximize a likelihood function?