Consider a new test for COVID-19. Assume that x% of the population has COVID-19 and that a test is y% sensitive (True Positive Rate) and z% specific (True Negative Rate). You pick the values of x, y, and z. Calculate the values for precision and recall and discuss how these relate to negative predictive validity, positive predictive validity, sensitivity, and specificity. Generate and post ROC and Precision Recall curves based on the values from your table as well as curve fitting. (You have three known values.: 1 from your table, 1 for all negative, and one for all positive. All other values are fictitious but might be generated by curve fitting.) Finally, estimate the probability for your values that an individual who tests positive for COVID-19 actually has COVID-19 (Bayes). Which metrics are most useful? Why?