[SOLVED] DATA622 Homework 1

Let’s use the Penguin dataset for our assignment. To learn more about the dataset, please visit:

https://allisonhorst.github.io/palmerpenguins/articles/intro.html

For this assignment, let us use ‘species’ as our outcome or the dependent variable.

1. Logistic Regression with a binary outcome.
   1. The penguin dataset has ‘species’ column. Please check how many categories

      you have in the species column. Conduct whatever data manipulation you need to do to be able to build a logistic regression with binary outcome. Please explain your reasoning behind your decision as you manipulate the outcome/dependent variable (species).

   2. Please make sure you are evaluating the independent variables appropriately in deciding which ones should be in the model.
   3. Provide variable interpretations in your model.
2. For your model from #1, please provide: AUC, Accuracy, TPR, FPR, TNR, FNR (20)
3. Multinomial Logistic Regression.
   1. Please fit it a multinomial logistic regression where your outcome variable is

      ‘species’.

   2. Please be sure to evaluate the independent variables appropriately to fit your

      best parsimonious model.

   3. Please be sure to interpret your variables in the model.
4. Extra credit: what would be some of the fit statistics you would want to evaluate for your model in question #3? Feel free to share whatever you can provide.