Homework 3 Solutions
by Ashley Van Reynolds (adapted from previous solutions)
Scoring:
20 Completion Points
10 Accuracy Points: #1, 4a, 5b,c, 6b,
... [Show More] c
1. Describe a real-world example of Type I and Type II error
(that was not mentioned in lecture or readings). Explain what
Type I and Type II error mean in the context of your example. 1
point completion, 1 point accuracy
In [3]: %matplotlib inline
import matplotlib as mat #import matplotlib
import seaborn as sns #import seaborn
import numpy as np #import numpy
import matplotlib.pyplot as plt
Homework 3 Solutions (Lecture 1 W21) 2/5/21, 1:03 AM
file:///Users/ana/Downloads/Assignment%20Solutions%2021W/Homew…0Solutions/Homework%203%20Solutions%20(Lecture%201%20W21).html Page 2 of 12
Type I error describes your false positive rate and thus is P(rejecting | True). \ Type II
error describes your false negative rate and thus is P(failing to reject | False).
Examples:
If a prenatal test comes back positive for Down Syndrome, when your fetus does not
have the disorder, that is a false positive (Type I error). If the prenatal test comes back
negative, but the fetus actually does have Down Syndrome, that is a false negative
(Type II error).
A false positive would be if a virus software on your computer incorrectly identifies a
harmless program as a malicious one. A false negative would be if the virus software
incorectly identifies a malicious program as a harmless one.
If a weapon is not detected by airport security, that's a false negative. If something is
identified as a weapon that is not a weapon, that's a false positive.
If a carbon monoxide detector beeps, even when the level of CO in the room is at a
healthy level, then the detector is giving a false positive. If the detector doesn't beep
when the level of CO is at a dangerously high level, then the detector is giving a false
negative.
Examples given in lecture 2.2 that do not receive credit:
Man being told he's pregnant (FP) or an obviously pregnant woman being told she's not
pregnant (FN)
A doctor viewing a kidney scan detects a tumor when there really isn't one (FP) or failing
to detect the tumor when there is one (FN) [Show Less]