Exam 3 Results

Wed, Apr 27

Use this key to translate your scores on the strip of paper:

Exam 1 = Exam 1 Score (out of 100)

Exam 2 = Exam 2 Score (out of 100)

HW+F4 = Homework + Final 4 score (out of 50)

Ex3 P1 = Exam 3 Part 1 take-home score (out of 100)

Ex3 P2 = Exam 3 Part 2 in-class score (out of 100)

Overall = Overall score (out of 100)

Exam 3 Review

Wed, Apr 20

Review for Exam 3:

• Part 1 Take-Home
• Part 2 In-Class
  • Logistic Regression
    • Output interpretation
    • Calculate probability
  • Big Data Analytics presentation
  • Python
  • Monte Carlo Simulation

Exam 3

[highlight color=”options: yellow, black”]Exam 3 on Mon, Apr 25[/highlight]

Part 1 Take Home Instructions:

• Software: Python or Sheets/Excel
• Data: S&P 500 Historical Returns
• Simulation: Stock Market Investment Returns
  • Randomly sample the S&P 500 Historical Returns to simulate stock market returns. Assume the historical return data are normally distributed.
  • Use a 40-year investment time horizon.
  • Start with a $10,000 balance.
  • Run a Monte Carlo simulation to generate 500 ending investment balances.
  • Find the mean and standard deviation of your ending balances.
  • Generate a histogram displaying the distribution of your ending balances.
• Print 1 Page with your (1) name, (2) section, (3) mean and standard deviation of ending balances, and (4) the histogram displaying the distribution of your ending balances. Bring your 1-page print out to your demo.
• Demo: [highlight color=”options: yellow, black”]You must demonstrate your Python program or Sheets/Excel worksheet to me in person[/highlight]. Be prepared to explain your simulation, answer questions and make (or allow me to make) modifications to your program or worksheet to test alternative parameters (e.g., different investment time horizon, different starting balance, different number of simulations, etc).
• Due Date: Yesterday. Just kidding. But in business clients always want everything yesterday and prefer to pay tomorrow. So get used to it. [highlight color=”options: yellow, black”]Demo with 1-page printout due on or before class on Apr 27[/highlight]. Demo as soon as you’re ready. This week is better than next week. I will have additional office hours TTh 2:30-4:30pm this week and next. Regular office hours are MW 12:30-2:30, 4:00-4:30pm.

Lesson 19: Random Sampling from a Normal Distribution

Previous | Next

Mon, Apr 18

Review:

• If you’re getting a syntax error, put this on the first line: [highlight color=”options: yellow, black”]#!/usr/bin/env python2.7[/highlight]
• Dice Game simulation

Presentation:

• Random Sampling from a Normal Distribution

• Simulate the distribution of ending balances for the dice game

Assignment:

• Exam 3 Part 1

Lesson 18: Monte Carlo Simulation

Previous | Next

Wed Apr 13

Review:

• Multiple Regression in Python

Presentation:

• Monte Carlo Simulation
• Resources
  • Tutorial
  • Basic Income vs Basic Job
  • Excel Solver with Monte Carlo Simulation
• Sample Code

Assignment:

• Create a simple Monte Carlo Simulation with Python
• Simulate a simple Dice Game:
  • Roll 2-7, lose
  • Roll 8-12, win
  • Start with $100
    • bet $1 per roll (win pays $1, loss takes $1 bet)
    • roll 100 times and find ending balance (could be negative)
    • simulate 100 outcomes (10,000 total rolls of the dice)
  • Report the mean, median, and standard deviation of your final balances
  • Repeat using 1,000 rolls per outcome (100,000 total rolls of the dice) and report the mean, median and standard deviation of final outcomes

[highlight color=”options: yellow, black”]Here’s some python code to help you get going (and plain text below): [/highlight]

import random

def rollDice():
roll = random.randint(1,6)
return roll

j = 0
rollcount = 0
rolls = []

while j < 100:
i = 0
j = j + 1

while i < 100:
i = i + 1
result = rollDice()
rolls.append(result)
rollcount += 1

import scipy
mr = scipy.mean(rolls)
print(“Total rolls = “)
print rollcount
print(“Mean roll = “)
print mr

Lesson 17: Multiple Regression in Python

Previous | Next

Mon, Apr 11

Review:

• Intro Statistical Computing with Python

Presentation:

• Survey on Residency Requirement: AdvStatsSurvey_Res04102016 (pdf)
• Schedule for the home stretch
  • Apr 11 – Multiple Regression in Python
  • Apr 13 – Intro to Monte Carlo Simulation
  • Apr 18 – Monte Carlo Simulation Project
  • Apr 20 – Review for Exam 3
  • Apr 25 – Exam 3
  • Apr 27 – Exam 3 Results and Final Exam Review
  • Finals Week

• Multiple Regression in Python
  • lesson16.py

Assignment:

• Complete problems 13.2 and 13.6 (data in Sheets) using Python.
• Create stand-alone programs

Lesson 16: Simple Statistical Computing with Python

Previous | Next

Wed, Apr 6

Review:

• Big Data Analytics
• Python Anywhere Setup and Hello World program
• Please take this Survey

Presentation:

• Computing basic statistics in Python
  • define and set variables
  • import scipy
  • calculate mean, median, standard deviation
  • run correlation
  • run linear regression
• Demo using Python to complete problem 12.1 from Lesson 3.
  • Step by step Python commands: PythonProblem12.1
• Sources
  • Sheppard, Kevin, 2014. “Introduction to Python for Econometrics, Statistics and Data Analysis”
  • Numpy and Scipy documentation

Assignment:

• Use Python to complete problems 12.2, 12.4, 12.6, 12.8 (p. 474-475) from Lesson 3 and Lesson 4.
• Print output (screen capture).
• Reading:
  • http://readwrite.com/2013/11/25/python-displacing-r-as-the-programming-language-for-data-science
  • http://www.infoworld.com/article/2951779/application-development/in-data-science-the-r-language-is-swallowing-python.html

Lesson 15: Big Data Analytics

Previous | Next

Review:

• Exam 2 Corrections Due Date Extension: submit by 1pm Thu, Apr 7
• Logistic Regression
• NCAA Final Four Predictions
  • AP Villanova vs Oklahoma
  • AP UNC vs Syracuse

Presentation:

• Big Data Analytics: AASA-TechConference-09212015 (Slides version)
• Key technologies in Data Science
  • SQL
  • R
  • SAS, SPSS, Minitab
  • Python
• Python Anywhere

Assignment:

• Create an account on https://www.pythonanywhere.com/
• Complete Python Lesson 1: Hello World
• Print “Hello World” + the last 4 digits of your PID, e.g., Print (“Hello World 0147”)