Workshop 2, Stats for AI

Author

Alberto Dorantes

Published

August 13, 2026

Abstract
This is an INDIVIDUAL workshop. In this workshop we practice with excercises related to Hypothesis testing.

1 General Directions for each workshop

You have to work on any front-end code editor (e.g. Google Colab, VSCode, etc) and create one Jupyter Notebook for each Workshop.

Name your W2-Notebook as “W2-StatsAI-YourFirstName-YourLastname”.

In your Workshop Notebook you have to respond to CHALLENGES/EXERCISES For each challenge, write down the directions before your solution.

  • It is STRONGLY RECOMMENDED that you write your OWN NOTES in this Jupyter Notebook; this will be very helpful to study for the final Evaluation.
    Once you finish your workshop, make sure that you RUN ALL CHUNKS. Then, render a Jupyter Notebook (.ipynb file)

You have to submit your .ipynb file to Canvas.

The material related to this workshop is covered in Chapters 6 and 7 my ebook. You can find my ebook at:

https://www.apradie.com/StatsBAbook/

2 CHALLENGE 1

Go to Chapter 6 and do the simulations starting in section 6.3.2. Follow what you get and try to understand the Central Limit Theorem with these simulations.

With your words, in 1-3 sentences, write what you learned from this exercise.

3 CHALLENGE 2

Collect real monthly data for Microsoft (MSFT) stock from Jan 2022 to Aug 2026.

You have to do the following:

Run a t-test to compare whether the mean return of MSFT is greater than zero.

To do a hypothesis test, we usually do the following steps:

  1. DEFINE THE VARIABLE OF STUDY.

  2. WRITE THE NULL AND THE ALTERNATIVE HYPOTHESIS.

  3. CALCULATE THE STANDARD ERROR, WHICH IS THE STANDARD DEVIATION OF THE VARIABLE OF STUDY.

  4. CALCULATE THE t-statistic (t-value). EXPLAIN/INTERPRET THE t-statistic.

  5. CALCULATE THE 95% CONFIDENCE INTERVAL OF THE SAMPLE MEAN

  6. WRITE YOUR CONCLUSION OF THE t-TEST

4 CHALLENGE 3

Considering monthly periods from Jan 2022 to Aug 2026, do a t-test to compare the mean return of ORACLE vs the mean return of NVIDIA.

Using HAND-WRITING, do the following:

Write down the Null and alternative hypothesis

Calculate the t-Statistic of the test (you can calculate the descriptive statistics in Python or Excel) and INTERPRET the result

What would be your conclusion of this test? Explain with your own words