In this lab you will…
You can find the repo for this lab here: https://classroom.github.com/a/WcqhI38O
Each person on the team should clone the repository and open a new project in RStudio.
We will use the following packages in this lab:
library(tidyverse)
library(tidymodels)
library(knitr)This dataset comes from Little et al. (2008). The data includes various measurements of dysphonia from 32 people, 24 with Parkinson’s disease (PD). Multiple measurements were taken per individual. The measurements we examine in this subset of the data include:
name: patient IDjitter: a measure of relative variation in fundamental frequencyshimmer: a measure of variation in amplitude (dB)PPE: pitch period entropyHNR: a ratio of total components vs. noise in the voice recordingstatus: health status (1 for PD, 0 for healthy)Use the code below to load the data sets into R.
parkinsons = read_csv("parkinsons.csv")Make sure we see all relevant code and output in the knitted PDF. If you use inline code, make sure we can still see the code used to derive that answer.
Write a narrative for each exercise.
All narrative should be written in full sentences, and visualizations should have clear title and axis labels.
For each individual, multiple repeated measurements were taken. For example, for individual S01, six repeated measurements were taken: _1, _2, etc.
Use the code below to remove the characters after the final underscore, i.e the individual measurement number for each individual. This will let you to group by participants.
park = parkinsons %>%
mutate(name = str_remove_all(name, "_[^_]+$"))What are the identification codes (names) of healthy individuals in the data set? Print your output as a nice kable table.
Let’s split the data into two disjoint sets: a training set and a test set. From the original data frame, remove 4 PD and 4 healthy individuals to be in your test set. For consistency, choose the 4 healthy individuals with the lowest ID number of their respective category, e.g. phon_R01_S07 ends with 07 (the lowest number of the healthy group) so they should be placed in the test data frame. Similarly, the lowest ID number for an individual with PD is S01 so phon_R01_S01 should also be placed in the test data frame.
Your train data frame should contain 147 rows and your test data frame should contain 48. In your code chunk, print the number of rows of each data frame.
Create a scatterplot of HNR vs shimmer. Color points by PD status. Make all labels informative. Comment on what you observe. Reminder: use all standard best plotting practices.
Build a main effects logistic regression model that predicts prob(PD) status from HNR, PPE, jitter and shimmer. Print your model tidy.
Which predictors are significant at the alpha = 0.05 level?
Edit the code chunk below, specifically renaming model_fit and test_data where appropriate. Uncomment and run to find the predicted probabilities of Parkinson’s disease in the test data frame.
# prediction = predict(model_fit, test_data, type = "prob")
# test_result = test_data %>%
# mutate(predicted_prob_pd = prediction$.pred_1)Next, create a new column that classifies an individual as having PD if the predicted probability is above 50%. Repeat with a decision boundary of 75% and 90%.
How many false positives do you have in each case? False negatives? If you were to use your model as a diagnostic tool for PD to decide if someone should undergo subsequent testing, which decision boundary would you prefer and why?
Note: your narrative should read, e.g.: “With a decision boundary of 50%, our model yields X false positives and Y false negatives. With a decision boundary of 75%…” etc.
There should only be one submission per team on Gradescope.
| Component | Points |
|---|---|
| Ex 1 | 6 |
| Ex 2 | 10 |
| Ex 3 | 7 |
| Ex 4 | 9 |
| Ex 5 | 12 |
| Workflow & formatting | 6 |