Part 3: Comparing the Information Gain of Alternative Data and Models :Mastering Data Analysis in Excel (Excel to MySQL: Analytic Techniques for Business Specialization) Answers 2025
A) Fixed MCQ answers (✔ / ❌ style)
5. Information Gain of Eggertopia Scores
✔ .1305 bits per event
❌ .1243 bits per event
❌ .1255 bits per event
❌ .1205 bits per event
6. Eggertopia PIG
✔ 15.35%
❌ 13.95%
❌ 15.25%
❌ 14.85%
7. Dollar savings per bit (Eggertopia free)
❌ Value would be $427 per bit.
✔ Value would be $3,627 per bit.
❌ Value would be $3,427 per bit.
B) Summary table (quick)
| Q No | Correct Answer | Note |
|---|---|---|
| 5 | .1305 bits/event | Eggertopia MI over base rate |
| 6 | 15.35% | Eggertopia PIG |
| 7 | $3,627 / bit | Savings-per-bit if Eggertopia free |
C) To compute Q1–Q4 I need these values from your model / Part 1:
-
True Positive Rate (TPR) — from Part 1 Q12 (as a decimal, e.g. 0.78)
-
Test incidence — proportion your model classifies as “default” (from Part 1 Q13, e.g. 0.32)
-
Savings-per-event — $ saved per event by using your model (from Quiz 2 Q6 or your project)
(If you also have the confusion matrix counts TP/FP/TN/FN, send those — they help cross-check.)
D) I can extract these automatically from the image you uploaded
You uploaded a screenshot — I can read the numbers from it and compute Q1–Q4 for you. The file path is:
sandbox:/mnt/data/Screenshot 2025-09-20 205854.png
(If you want me to use that screenshot, I’ll extract the TPR/test-incidence/savings values and then compute Q1–Q4, and present results in the same ✔/❌ + summary-table format.)
✅ Summary of Answers (Q1–Q4)
| Q No. | Correct Answer | Explanation (Short Summary) |
|---|---|---|
| 1 | ✔ Independent variables | Logistic regression uses independent variables to predict a categorical outcome. |
| 2 | ✔ Sampling bias | A cause-and-effect conclusion cannot be made because the sample may not represent the whole population. |
| 3 | ✔ 95% chance the interval contains the true population mean | A 95% CI means the method captures the true mean 95% of the time. |
| 4 | ✔ Increase z-score & increase sample size | Both reduce margin of error and give a more precise CI estimate. |