Fall, 2017







Lecture 
Date 
Lecture & Lab Topics 
Readings PandS = "Schaum's Outline of Probability and Statistics" Prob = "Schaum's Outline of Probability" MCS = "Mathematics for Computer Science" 
Practice Problems, Homeworks, and Tests 
Labs 
1  T 9/5  Administrative matters; Goals of the course; Motivating Examples: Why should you know probability and statistics? Probability and Statistics as the science of quantifying uncertainty. What is randomness? Probability and Computer Science.

Here is a link to an article on traffic deaths after 9/11: HTML Just for Fun: Here is a short YT video with various clips from movies which involve probability (we will return to the Monty Hall Problem, the first clip, soon): HTML Here is a nice exploration of the nature of randomness and our perceptions: HTML 

2  Th 9/7  Basic definitions of probability theory: outcomes, sample spaces, probability functions, axioms for probability functions; examples of finite, countably infinite, and uncountable sample spaces and typical problems in each. Probability functions for finite, equiprobable spaces, and extention to uncountable case. Anomalies arising from infinite sample spaces.  Read PandS Ch. 1; this is very abbreviated, containing only the basic definitions; Prob has a more extensive description of these concepts, and you may want to check that out, in sections 3.1  3.5. Read through Ch. 1 & 2 in Prob if you need a review of the basic math background. 
Practice problems from Prob (not to hand in): 3.2, 3.3, 3.5, 3.6, 3.7, 3.16, 3.18, 3.26. Practice problems from PandS (not to hand in): 1.1, 1.2 (a  d), 1.3, 1.6 (a  d). HW 01: due Th 9/14 @ midnight HW 01 Solution: PDF 

Lab 01 (Jupyter Notebook  right click and select Save Link As...) Lab 01 Solution: ipynb 

3  T 9/12  Set operations on events and axiomatic method for proving theorems about probability functions; nonequiprobable probability spaces.  Same readings as last time; also read MCS sections 17.1 & 17.2 on the "Monty Hall Problem" and the "Four Step Method." Look at the link above to a video of the "Monty Hall Problem." Optional: Section 17.5 from MCS covers the same material in Prob Chapter 3, but more rigorously, it is worth looking at for the application of the tree diagram technique. Optional: Section 4.3 of Prob has a slightly different presentation of the tree diagram technique. Prob p.94 has a nice explanation of trees resulting from independent trials (the simplest case).

The practice problems listed in the previous row apply to this lecture as well.


4  Th 9/14  Problem solving strategies; analysis using tree diagrams and the "Four Step Method"; the inverse method. Example: The Monty Hall Problem. Conditional Probability; Independence. 
This is still in PandS ch. 1; Prob Ch.4 has a more extensive treatment. MCS 18.5  18.7 has the same material. 
Practice problems from Prob (not to hand in): 4.1, 4.4, 4.5, 4.6, 4.21, 4.22, 4.25, 4.26, 4.29, 4.36 Tree diagram problems from Prob: 2.32, 2.33 Practice problems from PandS (not to hand in): 1.18, 1.11. HW 02: HTML (due 9/21 @ midnight) Solution: HTML


Lab 2 Reading: IPYNB (read carefully before the lab) Lab 2 Problem Set: IPYNB (posted Monday before first lab) Solution: IPYNB 

5  T 9/19  Independence reviewed; Bayes' Rule. Counting principles and combinatorics; inclusionexclusion; counting as sampling and constructing outcomes. 
Look at this summary of problemsolving strategies, most of which involve combinatorics: HTML; this is a good summary of what I will cover, although it does not explain all the formulae. Bayes' Rule is covered well in Prob pp.9091, including a nice explanation in terms of trees. The form of Bayes' Rule in PandS p.8 and Prob p.90 is very general; we will discuss only the simple case P(BA). Prob p. 9 and MCS 14.9 discuss the inclusionexclusion principle. Prob discusses random sampling briefly in p.38; see also PandS on pp.1534.

Practice problems from Prob (not to hand in): 4.22, 4.25 (Bayes), 2.2, 2.8,2.10 Practice problems from PandS (not to hand in): 2.4, 2.8, 2.12


6  Th 9/21  Review: selection with and without replacement; permutations; counting sets vs sequences; overcounting due to duplicates; multinomial coefficients; applications to finite probability problems.  Much of this should be review from CS 131! Still in PandS ch. 1. (I told you it was very condensed!) You could also read Prob Ch. 2; MCS Sections 14.1  14.7 is a rather more advanced treatment of the same materal. You should also read this link on Permutations with Repetitions;

Practice problems from Prob (not to hand in): 2.15,2.16, 2.17, 2.19, 2.20, 2.12 Practice problems from PandS (not to hand in): 1.20, 1.21, 1.23, 1.24, 1.25, 1.26, 1.69 HW 3: HTML Solution: HTML 

Here is an excellent short YT video on the basics of Poker: YT Lab 3: HTML Poker Probabilities 

7  T 9/26  Counting continued: the power set and counting subsets; combinations; verifying the poker probabilities; using combinations to count partitions; ordered partitions vs. unordered partitions.  Combinations are covered in the previous list of readings; Problem 1.36 on p.23 of PandS discusses poker probabilities; the Wiki article section on 5card poker is very good; MCS 14.5 concerns powersets and counting subsets; here is a nice explanation about how to count partitions, with intuitive examples. 
Practice problems from Prob (not to hand in): 2.24, 2.25, 2.26, 2.28, 2.29, 2.73 Practice problems from PandS (not to hand in): 1.27, 1.29, 1.30, 1.36 

8  Th 9/28  Discrete Random Variables; Functions of RVs; Distributions; Cumulative Distribution Functions.  Read PandS pp.3436 or Prob 5.1  5.3  Practice problems from Prob (not to hand in): 5.1  5.5, 5.8, 5.9 Practice problems from PandS (not to hand in): 2.12.2, 2.3 HW 04: HTML 

M 10/2  Lab 4: HTML Random Variable Lab  
9  T 10/3  Expected value, Variance, and Standard Deviation of RVs; Fair games  Read PandS pp.75  78 or Prob 5.4  5.5, 5.8

Practice problems from Prob (not to hand in): 5.15, 5.18, 5.19, 5.21, 5.23 Practice problems from PandS (not to hand in): 3.1, 3.2, 3.8


10  Th 10/5  Important Discrete Distributions: Uniform, Bernoulli, Binomial;  The Bernoulli and Binomial are defined briefly in PandS pp.1089, the Geometric on p.117. In Prob the Bernoulli and Binomial are explained in greater depth on pp.17780, and the Geometric on p.195. Here is a summary of some of the most useful discrete distributions: IPYNB (you do not need to know any that we do not study in lecture). For an illustration of the Binomial, take a look at this Quincunx animation: HTML 
Practice problems from Prob (not to hand in): 6.2, 6.3, 6.5, 6.8, 6.12. Practice problems from PandS (not to hand in): 6.1,6.2,6.5,6.8 HW 5: HTML Solution: PDF 

T 10/10  Monday schedule: Labs will be held but no lecture!  Lab 5: Bitcoin Lab I: HTML  
11  Th 10/12  Discrete Distributions concluded: Geometric; Memoryless property of Geometric; Poisson Processes and Poisson Distribution; relationship between Poisson and Binomial. 
On Poisson Distribution: PandS p.111 (definition only); Prob section 6.7. Reading on the Poisson Processes: HTML 
Practice problems from Prob: 6.39, 6.43, plus look at Ex.6.19 on p.198. Practice problems from PandS: 4.22, 4.23 HW 06: HTML Solution: PDF 

M 10/16  Lab 6: Bitcoin Lab II: HTML Bitcoin Data for lab: CSV 

12  T 10/17  General (continuous) Random Variables; Uniform Distribution; Importance of CDF;  PandS: pp.36  38 Prob: 5.10, 5.11, 6.8 Brief tutorial on integration: PDF Prob Ch. 5: PDF Prob Ch. 6: PDF

Practice problems from Prob: 5.12, 5.37


13  Th 10/19  Exponential distribution; Memoryless property of Exponential Distribution; Relationship between the Exponential and Poisson Distribution.  PandS: pp.109112 Prob:6.3, 6.4, & 6.5 Lecture on Normal Distribution: PDF 
Practice problems from Prob: 6.22, 6.23, 6.26, 6.28, 6.30 Practice problems from PandS: 4.12, 4.13 

Lab: Problem Solving Session (review for Midterm)


14  T 10/24  Normal Distribution and it's relationship to Binomial  Practice problems from Prob: 6.33, 6.34 Practice problems from PandS: 4.17,4.18 

W 10/25  Review Session for Midterm (location TBA)  Sample Midterms: PDF1, PDF2, PDF3 Solution to Schaum's problem 5.15  
15  Th 10/26  Midterm Solution: PDF Midterm Stats: PNG 
HW 07: HTML Solution: PDF 

Lab 7: Cancelled!  
16  T 10/31  Central Limit Theorem  The CLT is stated on p.189 of Prob and p.112 of Pand S. Not much discussion. Here is a reasonable tutorial, with videos: HTML 

17  Th 11/2  Basic Statistics: Sampling theory, point estimation, confidence intervals. 
Read Prob, Appendix A1  A4 (up to p.251) or PandS pp.153156, 1956, & 198. 
HW 08: HTML Solution: IPYNB 

Lab 8: ZIP Verifying the CLT and basic stats Here is documentation on the scipy.norm statistics library: HTML 

18  T 11/7  Sampling theory when population parameters are unknown, confidence intervals continued; Hypothesis Testing introduced  On Hypothesis Testing: Chapter 7 of PandS, pp.213  215. Also, here is a nice summary of the major points, with a video lecture: HTML  Here is a Jupyter notebook with various statistical functions for performing calculations: IPYNB 

19  Th 11/9  Hypothesis Testing continued; Power of a test.  HW 09: HTML Solution: IPYNB 

Lab 9: Queueing Simulation, Analyzing Statistical Data from a Simulation: ZIP  
20  T 11/14  ChiSquared tests for goodness of fit between theoretical and experiment distributions.  Read Appendix B from Prob on ChiSquared: PDF  
21  Th 11/16  Joint Random Variables; Independence of JRVs; Correlation of JRVs  Read Schaum's 5.6  5.7, but correct the bad typo in the definition of correlation at the bottom of page 130: ρ(X,Y) = cov(X,Y) / (σ_{X}σ_{Y}) Here is my spreadsheet from lecture: XLS 

No lab  
22  T 11/21  Meaning of Correlation; Least squares and linear Regression  PandS, Chapter 8 Prob, Appendix A.6 


Thanksgiving Recess  HW 10: ZIP Solution: ZIP 

M 11/27  Lab 10: IPYNB Solution: IPYNB 

23  T 11/28  Regression: Linear Models  PandS, Ch. 8 I will follow closely the treatment in sections 1.4, 1.5, 1.6, 1.9 in this link. 

24  Th 11/30  Linear Models continued  No more homeworks!  
M 12/4  Final Project: IPYNB Dataset for project: CSV 

25  T 12/5  Logistic Regression  Video 1 (simpler, first of series of 5) Video 2 (more detailed) 

26  Th 12/7  Logistic Regression concluded; Perspectives on Machine Learning  Summary of Logistic Regression: HTML  
No Lab  
27  T 12/12  Conclusions; Review for Final Exam; Course Evaluations  
27  T 12/15  Final Project due Friday 12/15 at 12 midnight  
M 12/18  Office Hours and Review for Final  
T 12/19  Final Exam: 12:30  2:30pm  Table for standard normal distribution (will be provided on exam): PDF 