Lecture 18: Central Limit Theorem. Lisa Yan August 6, 2018
|
|
- Andrew Campbell
- 5 years ago
- Views:
Transcription
1 Lecture 18: Central Limit Theorem Lisa Yan August 6, 2018
2 Announcements PS5 due today Pain poll PS6 out today Due next Monday 8/13 (1:30pm) (will not be accepted after Wed 8/15) Programming part: Java, C, C++, or Python Lecture Notes published for all material 2
3 Summary from last time! " $ & ' ( for all $ > 0 if " 0 (Markov s inequality)! " -. /0 1 0 for all. > 0 (Chebyshev s inequality) if 2 " = -, Var " = 4 5! " $ /0 / 0 6( 0 for all $ > 0 (One-sided Chebyshev s (and others) if 2 " = 0, Var " = 4 5 inequality) 2 7 " 7 2 " if for all x (Jensen s inequality) 3
4 Bounds on sample mean Consider a sample of n i.i.d random variables X 1, X 2,... X n, where X k has distribution F with E[X k ] = µ and Var(X k ) = s 2 Sample mean!" = $ % % '($ " ' )!" = *, Var!" = +, % 4
5 Summary from last time Weak Law of Large Numbers lim ' () + - = 0 $ & à probability goes to zero in the limit à Will still have a deviation in the limit (at most -) à We have an infinite number of non-zero probabilities Strong Law of Large Numbers ' lim () = + = 1 $ & à Implies WLLN à After some number n, ' () + - = 0 à We have a finite number of non-zero probabilities 5
6 Where are we going with all of this? As engineers, we want to: 1. Model things that can be random. 2. Find average values of situations that let us make good decisions. We perform multiple experiments to help us with engineering. Counts and averages are both sums of RVs. (RVs) (expectation) Weeks 4 and 5: Learning how to model multiple RVs This week (week 6): Finding expectation of complex situations Defining sampling and using existing data Mathematical bounds w.r.t modeling the average Week 7: bringing it all together Sampling + average = Central Limit Theorem Samples + modeling = finding the best model parameters given data 6
7 Goals for today Central Limit Theorem! Confidence intervals, re-defined Estimates for sums of IID RVs Introduction to Parameter estimation 7
8 Central Limit Theorem If! " # = %, Var " # = ' (, Sample means of IID RVs are normally distributed. / )" = 1 +, " # 1 %, '( + #-. Sums of IID RVs are normally distributed. /, " # 1 +%, +' ( #-. 8
9 CLT explains a lot!~bin # = 200, ( = P(X = x) x Normal approximation to Binomial when np(1-p) > 10 9
10 A short history of the CLT : CLT for X ~ Ber(1/2) postulated by Abraham de Moivre : Pierre-Simon Laplace extends de Moivre s work to approximating Bin(n,p) with Normal : Alexandr Lyapunov provides precise definition and rigorous proof of CLT 2012: Drake releases Started from the Bottom As of July 4 th, he has a total of 190 songs Mean quality of subsamples of songs is normally distributed (thanks to the Central Limit Theorem) 10
11 Implications of CLT Anything that is a sum/average of independent random variables is normal meaning in real life, many things are normally distributed: Exam scores: sum of individual problems Movie ratings: averages of independent viewer scores Polling: Ask 100 people if they will vote for candidate X p 1 = # yes /100 Sum of Bernoulli RVs (each person independently says yes w.p. p) Repeat this process with different groups to get p 1,, p n (different sample statistics) Normal distribution over sample means p k Confidence interval: How likely is it that an estimate for true p is close? 11
12 No more convolution when!! For independent discrete random variables X and Y, and Z = X + Y, $ % & = ( ) + + = & =, $. (0)$ 2 (& 0) =, $. (& 5)$ 2 (5) - 4 p Z is defined as the convolution of p X and p Y. Sums of IID RVs are normally distributed. 9, ) 6 ;!<,!>? 678 (caveat: must be independent and identically distributed) 12
13 CLT vs. Water speed stat - +! = "/,! " 9 :, ;< 6 Water Pokemon speed statistic,! " : #! " = 65, Var! " = Water Pokemon Let Y = average of 50 Water Pokemon. What is the distribution of Y? Solution: * = +! =, - "/, -! " # * = #! " = 65 Var * = Var = 2,3 23 = 10.2 * 9 65,10.2! 13
14 Confidence interval, fully defined Consider a sample of IID RVs X 1, X 2,, X n, where n is large, "# = % & ()% & # (, Var "# = *+ &,- = % & &.% ()% # ( "# -, SE = 2 & def For large n, a 100(1 α)% confidence interval of 3 is defined as: where Φ(4 5/- ) = 1 (>/2). ex: 95% confidence interval: "# 4 5/- 7 SE, "# + 4 5/- 7 SE "# ± 4 5/- 7 (SE) "# ± 1.96(SE) > = 0.05, >/2 = Φ 4 5/- = 0.975, 4 5/- = 1.96 In other words: 95% of time that we sample, true 3 will be in interval "# ± 1.96(SE) NOT: 3 is 95% likely to be in this particular interval. 14!
15 SE with confidence interval Idle CPUs are the bane of our existence. A large company monitors 225 computers (single CPU) for idle hours in order to estimate the average number of idle hours per CPU. Sample had!" = 11.6 hrs, ' ( = hrs 2 (so ' = 4.1 hrs) Give the 90% confidence interval for the estimate of +, mean idle hrs/cpu. Solution:, = 0.10,,/2 = 0.05 Φ 3 4/( = 0.95, 3 4/( = Standard error, SE = 8 = :.; = :.; 9 ((< ;< 90% confidence interval:!" ± 3 4/( (SE) à 11.6 ± :.; 11.15, Interpret: 90% of time that such an interval is computed, the true + is it. ;<! 15
16 Break Attendance: tinyurl.com/cs109summer
17 Sampling distribution CLT tells us sampling distribution of!" is approx. normal when n is large. large n: n > 30 (larger is better): n > 100 We can also use CLT to decide values of n for good confidence intervals 17
18 Estimating clock running time * '( = A,-) (, / Want to test runtime of algorithm. Run algorithm n times and measure average time Know variance is! " = 4 sec 2 Want to estimate mean runtime: % = & How large should n be so that P(average time is within 0.5 of t) = 0.95? Solution: '( = ) *,-) * (, / &, 1 * (by Central Limit Theorem) 2 & 0.5 '( & = 0.95 = '( & 0.5 ( '( &) / 0, 1 * = 2 <.= * " > <.= * " > = ' ( & ~/(0,1) (SD: " * )! 18
19 Estimating clock running time 2 -. = 1 > D EFG. E I %,!" > Want to test runtime of algorithm. Run algorithm n times and measure average time Know variance is! " = 4 sec 2 Want to estimate mean runtime: % = & How large should n be so that P(average time is within 0.5 of t) = 0.95? Solution: ' & & = ' & 0.5 = ' = Φ 2 6 Φ 2 6 Φ = Φ 2 6 = Solve for n*: 1 Φ = G 2 EFG 2. E I &, 6 2 = 2Φ 2 6 What say you, Chebyshev? ' -. & ' -. & 0.5 6/ B 16/> 0.05 > 320 " " 1 = 0.95 = = 1.96 > = 7.84 " = 62 6 Thanks for playing, Pafnuty!! 19
20 The power of the central limit theorem Sample means of IID RVs are normally distributed. *!" = 1 % & '() Can estimate confidence interval Can give probabilities on sample mean " ', -, /0 % Sums of IID RVs are normally distributed. * & " ', %-, %/ 0 Can estimate any sum of IID RVs (if -, / 0 known) '() our next goal 20
21 Crashing website : 7 # 8 < =>, =@ A 890 Let X = number of visitors to a website/minute: X~Poi(100) The server crashes if 120 requests/minute. What is P(crash in next minute)? Solution 1: (exact, using Poisson) Solution 2: (using CLT) Recall: sum of IID Poissons is Poisson CLT says sums are normally distributed (as n à ) Let: : Poi(100)~ 7 B90 Poi(100/=) " # 120 = 7 B90A1 " # 120 " ) = "./ / C D / B Using CLT on discrete sum à continuity correction Attempted solution 3: (using one-sided Chebyshev) " # 120 = " # J # A + L A = A = 0.2 E! = 1 Φ ! 21
22 Summary: Approximations Poisson: Poisson = sum of IID Poissons where n is arbitrarily large Sum of IID RVs with known E[X], Var(X) continuity correction possible continuity correction Normal: (by Central Limit Theorem) Binomial: n trials P(success) = p, Independent trials continuity correction Normal: p medium, np(1 p)>10 Independent trials Poisson: p < 0.05, n > 20 AND/OR slight dependence! 22
23 Another dice game You will roll 10 6-sided dice (X 1, X 2,, X 10 ) Let: X = total value of all 10 dice = X 1 + X X 10 Win if: X 25 or X 45 What is P(win)? Roll!!!!!!! And now the truth (according to CLT)! 23
24 Another dice game You will roll 10 6-sided dice (X 1, X 2,, X 10 ) Let: X = total value of all 10 dice = X 1 + X X 10 Win if: X 25 or X 45 What is P(win)? Solution: WTF:! " 25 or " 45 = 1!(25 < " < 45) By CLT, " = 2 /01 " / approx. 3 45, 47 8 where E " / = 5 = 3.5,Var " / = 7 8 = <= 18 " = 35, 10 35/12 = 350/12 1! 25 < " < 45 1! 25.5 " 44.5 = 1! 8=.=B<= <=C/18 DB<= EE.=B<= <=C/18 <=C/18 1 2Φ = continuity correction! 24
25 Another dice game You will roll 10 6-sided dice (X 1, X 2,, X 10 ) Let: X = total value of all 10 dice = X 1 + X X 10 Win if: X 25 or X 45 What is P(win)? Solution:! " 25 or " 45 = 1!(25 < " < 45) = ! 25
26 Finding Statistics What do you know? Distribution with parameters Sample of population RVs Bootstrapping What statistic can you find? Anything your heart desires Estimates of E[X] or Var(X), confidence interval, p-values E[X] or Var(X) Sum of IID RVs? CLT Probability bounds 26
27 Now What do you know? Distribution with parameters RVs What statistic can you find? Anything your heart desires What if you don t know the parameters of your distribution? 27
28 Parameter estimation def parametric model probability distribution that has parameters θ. Distribution Model Parameters θ Ber(p) Poi(λ) Multinomial(p 1, p 2,, p m ) Unif(α, β) Normal(µ, σ 2 ) Bernoulli Poisson Multinomial Uniform Normal! = #! = $! = # %, # ',, # )! = *, +! =,, - ' etc. Note θ can be a vector of parameters!! 28
29 Why do we care? def Parameter estimation 1. We observe data that has a known model with unknown parameter ". 2. Use data to estimate model parameters as!". Note: estimators!" are RVs estimating parameter " (e.g., sample mean:!" = $% estimates population mean " = &) Point estimate of parameter: Find the best single value for parameter estimate (as opposed to distribution of!") Better understand of process producing data Future predictions based on model Simulation of future processes Machine Learning: Learn model of system Do stuff with model 29
30 Estimator bias def bias of estimator:! "# # unbiased estimator: bias = 0 Example Sample Mean Estimator: '( = ) *,-) Bias:! '( =. Example 2 Sample Variance Estimator: / 0 = ) * *1) 2-) Bias:! / 0 = 3 0 * (, Estimating:. (unbiased estimator) ( 2 '( 0 Estimating: 3 0 (unbiased estimator)! 30
31 Estimator consistency def consistent estimator: lim $ & ' () ) <, = 1 for, > 0 As we get more data, the estimate () should deviate from the true ) by at most a small amount. Actually known as weak consistency Example Sample Mean Estimator: 12 = 3 $ 563 $ 2 5 Estimating: 7 By strong law of large numbers (SLLN): ' lim $ & 12 = 7 = 1 Implying weak law of large numbers (WLLN): lim $ & ' 12 7, = 0 for, > 0 Equivalently: lim ' 12 7 <, = 1 for, > 0 $ & (consistent estimator)! 31
32 Sample variance consistency def biased estimator:! "# # 0 def consistent estimator: lim - *, "# # < / for / > 0 As 1, estimate "# should deviate from true # by at most a small amount. Are the below two estimators for variance 3 4 biased? Consistent? = 7 * *87 :;7 < : =< 4 (sample variance) unbiased:! 5 4 = 3 4 consistent: 5 4 = * *87 7 * :;7 * < 4 : =< 4, 7 - lim *, * :;7 * < 4 : =! < 4 = 1, - lim =< 4 4 = 1 (by SLLN, < : i.i.d.) *, 2. A = 7 * * :;7 < : < 4! biased:! A =! *87 * 54 = *87 * 34 *87 consistent: for similar reasons as above #1, where now lim = 1 *, * 32
Central Theorems Chris Piech CS109, Stanford University
Central Theorems Chris Piech CS109, Stanford University Silence!! And now a moment of silence......before we present......a beautiful result of probability theory! Four Prototypical Trajectories Central
More informationLecture 13: Covariance. Lisa Yan July 25, 2018
Lecture 13: Covariance Lisa Yan July 25, 2018 Announcements Hooray midterm Grades (hopefully) by Monday Problem Set #3 Should be graded by Monday as well (instead of Friday) Quick note about Piazza 2 Goals
More informationLecture 10: Normal RV. Lisa Yan July 18, 2018
Lecture 10: Normal RV Lisa Yan July 18, 2018 Announcements Midterm next Tuesday Practice midterm, solutions out on website SCPD students: fill out Google form by today Covers up to and including Friday
More informationLecture 08: Poisson and More. Lisa Yan July 13, 2018
Lecture 08: Poisson and More Lisa Yan July 13, 2018 Announcements PS1: Grades out later today Solutions out after class today PS2 due today PS3 out today (due next Friday 7/20) 2 Midterm announcement Tuesday,
More informationLecture Notes 5 Convergence and Limit Theorems. Convergence with Probability 1. Convergence in Mean Square. Convergence in Probability, WLLN
Lecture Notes 5 Convergence and Limit Theorems Motivation Convergence with Probability Convergence in Mean Square Convergence in Probability, WLLN Convergence in Distribution, CLT EE 278: Convergence and
More informationFundamental Tools - Probability Theory IV
Fundamental Tools - Probability Theory IV MSc Financial Mathematics The University of Warwick October 1, 2015 MSc Financial Mathematics Fundamental Tools - Probability Theory IV 1 / 14 Model-independent
More informationOverview. Confidence Intervals Sampling and Opinion Polls Error Correcting Codes Number of Pet Unicorns in Ireland
Overview Confidence Intervals Sampling and Opinion Polls Error Correcting Codes Number of Pet Unicorns in Ireland Confidence Intervals When a random variable lies in an interval a X b with a specified
More informationPoisson Chris Piech CS109, Stanford University. Piech, CS106A, Stanford University
Poisson Chris Piech CS109, Stanford University Piech, CS106A, Stanford University Probability for Extreme Weather? Piech, CS106A, Stanford University Four Prototypical Trajectories Review Binomial Random
More informationBinomial in the Limit
Lisa Yan CS 09 Lecture Notes #8 July 3, 208 The Poisson Distribution and other Discrete Distributions Based on a chapter by Chris Piech Binomial in the Limit Recall the example of sending a bit string
More informationLecture 13. Poisson Distribution. Text: A Course in Probability by Weiss 5.5. STAT 225 Introduction to Probability Models February 16, 2014
Lecture 13 Text: A Course in Probability by Weiss 5.5 STAT 225 Introduction to Probability Models February 16, 2014 Whitney Huang Purdue University 13.1 Agenda 1 2 3 13.2 Review So far, we have seen discrete
More informationCSE 312: Foundations of Computing II Quiz Section #10: Review Questions for Final Exam (solutions)
CSE 312: Foundations of Computing II Quiz Section #10: Review Questions for Final Exam (solutions) 1. (Confidence Intervals, CLT) Let X 1,..., X n be iid with unknown mean θ and known variance σ 2. Assume
More informationCS145: Probability & Computing
CS45: Probability & Computing Lecture 5: Concentration Inequalities, Law of Large Numbers, Central Limit Theorem Instructor: Eli Upfal Brown University Computer Science Figure credits: Bertsekas & Tsitsiklis,
More informationthe law of large numbers & the CLT
the law of large numbers & the CLT Probability/Density 0.000 0.005 0.010 0.015 0.020 n = 4 0.0 0.2 0.4 0.6 0.8 1.0 x-bar 1 sums of random variables If X,Y are independent, what is the distribution of Z
More informationCOMP2610/COMP Information Theory
COMP2610/COMP6261 - Information Theory Lecture 9: Probabilistic Inequalities Mark Reid and Aditya Menon Research School of Computer Science The Australian National University August 19th, 2014 Mark Reid
More informationCOMPSCI 240: Reasoning Under Uncertainty
COMPSCI 240: Reasoning Under Uncertainty Andrew Lan and Nic Herndon University of Massachusetts at Amherst Spring 2019 Lecture 20: Central limit theorem & The strong law of large numbers Markov and Chebyshev
More informationMAS113 Introduction to Probability and Statistics
MAS113 Introduction to Probability and Statistics School of Mathematics and Statistics, University of Sheffield 2018 19 Identically distributed Suppose we have n random variables X 1, X 2,..., X n. Identically
More information1 Basic continuous random variable problems
Name M362K Final Here are problems concerning material from Chapters 5 and 6. To review the other chapters, look over previous practice sheets for the two exams, previous quizzes, previous homeworks and
More information1. Let X be a random variable with probability density function. 1 x < f(x) = 0 otherwise
Name M36K Final. Let X be a random variable with probability density function { /x x < f(x = 0 otherwise Compute the following. You can leave your answers in integral form. (a ( points Find F X (t = P
More informationIntroduction to Statistical Data Analysis Lecture 4: Sampling
Introduction to Statistical Data Analysis Lecture 4: Sampling James V. Lambers Department of Mathematics The University of Southern Mississippi James V. Lambers Statistical Data Analysis 1 / 30 Introduction
More informationIntroduction to Probability
LECTURE NOTES Course 6.041-6.431 M.I.T. FALL 2000 Introduction to Probability Dimitri P. Bertsekas and John N. Tsitsiklis Professors of Electrical Engineering and Computer Science Massachusetts Institute
More informationMidterm #1. Lecture 10: Joint Distributions and the Law of Large Numbers. Joint Distributions - Example, cont. Joint Distributions - Example
Midterm #1 Midterm 1 Lecture 10: and the Law of Large Numbers Statistics 104 Colin Rundel February 0, 01 Exam will be passed back at the end of class Exam was hard, on the whole the class did well: Mean:
More information18.175: Lecture 13 Infinite divisibility and Lévy processes
18.175 Lecture 13 18.175: Lecture 13 Infinite divisibility and Lévy processes Scott Sheffield MIT Outline Poisson random variable convergence Extend CLT idea to stable random variables Infinite divisibility
More informationCS 5014: Research Methods in Computer Science. Bernoulli Distribution. Binomial Distribution. Poisson Distribution. Clifford A. Shaffer.
Department of Computer Science Virginia Tech Blacksburg, Virginia Copyright c 2015 by Clifford A. Shaffer Computer Science Title page Computer Science Clifford A. Shaffer Fall 2015 Clifford A. Shaffer
More informationSTATS 200: Introduction to Statistical Inference. Lecture 29: Course review
STATS 200: Introduction to Statistical Inference Lecture 29: Course review Course review We started in Lecture 1 with a fundamental assumption: Data is a realization of a random process. The goal throughout
More informationDebugging Intuition. How to calculate the probability of at least k successes in n trials?
How to calculate the probability of at least k successes in n trials? X is number of successes in n trials each with probability p # ways to choose slots for success Correct: Debugging Intuition P (X k)
More informationWeek 6, 9/24/12-9/28/12, Notes: Bernoulli, Binomial, Hypergeometric, and Poisson Random Variables
Week 6, 9/24/12-9/28/12, Notes: Bernoulli, Binomial, Hypergeometric, and Poisson Random Variables 1 Monday 9/24/12 on Bernoulli and Binomial R.V.s We are now discussing discrete random variables that have
More informationBernoulli and Binomial
Bernoulli and Binomial Will Monroe July 1, 217 image: Antoine Taveneaux with materials by Mehran Sahami and Chris Piech Announcements: Problem Set 2 Due this Wednesday, 7/12, at 12:3pm (before class).
More informationthe law of large numbers & the CLT
the law of large numbers & the CLT Probability/Density 0.000 0.005 0.010 0.015 0.020 n = 4 x-bar 1 sums of random variables If X,Y are independent, what is the distribution of Z = X + Y? Discrete case:
More information2. Suppose (X, Y ) is a pair of random variables uniformly distributed over the triangle with vertices (0, 0), (2, 0), (2, 1).
Name M362K Final Exam Instructions: Show all of your work. You do not have to simplify your answers. No calculators allowed. There is a table of formulae on the last page. 1. Suppose X 1,..., X 1 are independent
More informationSS257a Midterm Exam Monday Oct 27 th 2008, 6:30-9:30 PM Talbot College 342 and 343. You may use simple, non-programmable scientific calculators.
SS657a Midterm Exam, October 7 th 008 pg. SS57a Midterm Exam Monday Oct 7 th 008, 6:30-9:30 PM Talbot College 34 and 343 You may use simple, non-programmable scientific calculators. This exam has 5 questions
More informationIntroduction and Overview STAT 421, SP Course Instructor
Introduction and Overview STAT 421, SP 212 Prof. Prem K. Goel Mon, Wed, Fri 3:3PM 4:48PM Postle Hall 118 Course Instructor Prof. Goel, Prem E mail: goel.1@osu.edu Office: CH 24C (Cockins Hall) Phone: 614
More informationWeek 9 The Central Limit Theorem and Estimation Concepts
Week 9 and Estimation Concepts Week 9 and Estimation Concepts Week 9 Objectives 1 The Law of Large Numbers and the concept of consistency of averages are introduced. The condition of existence of the population
More information1 Basic continuous random variable problems
Name M362K Final Here are problems concerning material from Chapters 5 and 6. To review the other chapters, look over previous practice sheets for the two exams, previous quizzes, previous homeworks and
More informationDiscrete Distributions
Discrete Distributions STA 281 Fall 2011 1 Introduction Previously we defined a random variable to be an experiment with numerical outcomes. Often different random variables are related in that they have
More informationLecture 7: Chapter 7. Sums of Random Variables and Long-Term Averages
Lecture 7: Chapter 7. Sums of Random Variables and Long-Term Averages ELEC206 Probability and Random Processes, Fall 2014 Gil-Jin Jang gjang@knu.ac.kr School of EE, KNU page 1 / 15 Chapter 7. Sums of Random
More informationThe central limit theorem
14 The central limit theorem The central limit theorem is a refinement of the law of large numbers For a large number of independent identically distributed random variables X 1,,X n, with finite variance,
More information6.1 Moment Generating and Characteristic Functions
Chapter 6 Limit Theorems The power statistics can mostly be seen when there is a large collection of data points and we are interested in understanding the macro state of the system, e.g., the average,
More informationMidterm 2 Review. CS70 Summer Lecture 6D. David Dinh 28 July UC Berkeley
Midterm 2 Review CS70 Summer 2016 - Lecture 6D David Dinh 28 July 2016 UC Berkeley Midterm 2: Format 8 questions, 190 points, 110 minutes (same as MT1). Two pages (one double-sided sheet) of handwritten
More informationLecture 10: Probability distributions TUESDAY, FEBRUARY 19, 2019
Lecture 10: Probability distributions DANIEL WELLER TUESDAY, FEBRUARY 19, 2019 Agenda What is probability? (again) Describing probabilities (distributions) Understanding probabilities (expectation) Partial
More informationDisjointness and Additivity
Midterm 2: Format Midterm 2 Review CS70 Summer 2016 - Lecture 6D David Dinh 28 July 2016 UC Berkeley 8 questions, 190 points, 110 minutes (same as MT1). Two pages (one double-sided sheet) of handwritten
More informationEECS 126 Probability and Random Processes University of California, Berkeley: Spring 2017 Kannan Ramchandran March 21, 2017.
EECS 126 Probability and Random Processes University of California, Berkeley: Spring 2017 Kannan Ramchandran March 21, 2017 Midterm Exam 2 Last name First name SID Name of student on your left: Name of
More informationCS 361: Probability & Statistics
February 19, 2018 CS 361: Probability & Statistics Random variables Markov s inequality This theorem says that for any random variable X and any value a, we have A random variable is unlikely to have an
More informationX = X X n, + X 2
CS 70 Discrete Mathematics for CS Fall 2003 Wagner Lecture 22 Variance Question: At each time step, I flip a fair coin. If it comes up Heads, I walk one step to the right; if it comes up Tails, I walk
More informationProbability Models of Information Exchange on Networks Lecture 1
Probability Models of Information Exchange on Networks Lecture 1 Elchanan Mossel UC Berkeley All Rights Reserved Motivating Questions How are collective decisions made by: people / computational agents
More informationPractice Problem - Skewness of Bernoulli Random Variable. Lecture 7: Joint Distributions and the Law of Large Numbers. Joint Distributions - Example
A little more E(X Practice Problem - Skewness of Bernoulli Random Variable Lecture 7: and the Law of Large Numbers Sta30/Mth30 Colin Rundel February 7, 014 Let X Bern(p We have shown that E(X = p Var(X
More information3 Multiple Discrete Random Variables
3 Multiple Discrete Random Variables 3.1 Joint densities Suppose we have a probability space (Ω, F,P) and now we have two discrete random variables X and Y on it. They have probability mass functions f
More informationCentral Limit Theorem and the Law of Large Numbers Class 6, Jeremy Orloff and Jonathan Bloom
Central Limit Theorem and the Law of Large Numbers Class 6, 8.5 Jeremy Orloff and Jonathan Bloom Learning Goals. Understand the statement of the law of large numbers. 2. Understand the statement of the
More informationExample continued. Math 425 Intro to Probability Lecture 37. Example continued. Example
continued : Coin tossing Math 425 Intro to Probability Lecture 37 Kenneth Harris kaharri@umich.edu Department of Mathematics University of Michigan April 8, 2009 Consider a Bernoulli trials process with
More informationS n = x + X 1 + X X n.
0 Lecture 0 0. Gambler Ruin Problem Let X be a payoff if a coin toss game such that P(X = ) = P(X = ) = /2. Suppose you start with x dollars and play the game n times. Let X,X 2,...,X n be payoffs in each
More informationSTAT 135 Lab 5 Bootstrapping and Hypothesis Testing
STAT 135 Lab 5 Bootstrapping and Hypothesis Testing Rebecca Barter March 2, 2015 The Bootstrap Bootstrap Suppose that we are interested in estimating a parameter θ from some population with members x 1,...,
More informationCSE 103 Homework 8: Solutions November 30, var(x) = np(1 p) = P r( X ) 0.95 P r( X ) 0.
() () a. X is a binomial distribution with n = 000, p = /6 b. The expected value, variance, and standard deviation of X is: E(X) = np = 000 = 000 6 var(x) = np( p) = 000 5 6 666 stdev(x) = np( p) = 000
More informationDiscrete Mathematics for CS Spring 2007 Luca Trevisan Lecture 20
CS 70 Discrete Mathematics for CS Spring 2007 Luca Trevisan Lecture 20 Today we shall discuss a measure of how close a random variable tends to be to its expectation. But first we need to see how to compute
More informationDiscrete Random Variables
Discrete Random Variables An Undergraduate Introduction to Financial Mathematics J. Robert Buchanan Introduction The markets can be thought of as a complex interaction of a large number of random processes,
More informationProbability Models. 4. What is the definition of the expectation of a discrete random variable?
1 Probability Models The list of questions below is provided in order to help you to prepare for the test and exam. It reflects only the theoretical part of the course. You should expect the questions
More information18.175: Lecture 17 Poisson random variables
18.175: Lecture 17 Poisson random variables Scott Sheffield MIT 1 Outline More on random walks and local CLT Poisson random variable convergence Extend CLT idea to stable random variables 2 Outline More
More informationa zoo of (discrete) random variables
discrete uniform random variables A discrete random variable X equally liely to tae any (integer) value between integers a and b, inclusive, is uniform. Notation: X ~ Unif(a,b) a zoo of (discrete) random
More informationM(t) = 1 t. (1 t), 6 M (0) = 20 P (95. X i 110) i=1
Math 66/566 - Midterm Solutions NOTE: These solutions are for both the 66 and 566 exam. The problems are the same until questions and 5. 1. The moment generating function of a random variable X is M(t)
More informationsuccess and failure independent from one trial to the next?
, section 8.4 The Binomial Distribution Notes by Tim Pilachowski Definition of Bernoulli trials which make up a binomial experiment: The number of trials in an experiment is fixed. There are exactly two
More informationThe Normal Distribution
The Normal Distribution image: Etsy Will Monroe July 19, 017 with materials by Mehran Sahami and Chris Piech Announcements: Midterm A week from yesterday: Tuesday, July 5, 7:00-9:00pm Building 30-105 One
More informationClass 26: review for final exam 18.05, Spring 2014
Probability Class 26: review for final eam 8.05, Spring 204 Counting Sets Inclusion-eclusion principle Rule of product (multiplication rule) Permutation and combinations Basics Outcome, sample space, event
More informationSampling Distributions
Sampling Error As you may remember from the first lecture, samples provide incomplete information about the population In particular, a statistic (e.g., M, s) computed on any particular sample drawn from
More informationIndependent random variables
Will Monroe July 4, 017 with materials by Mehran Sahami and Chris Piech Independent random variables Announcements: Midterm Tomorrow! Tuesday, July 5, 7:00-9:00pm Building 30-105 (main quad, Geology Corner)
More informationStatistics 1B. Statistics 1B 1 (1 1)
0. Statistics 1B Statistics 1B 1 (1 1) 0. Lecture 1. Introduction and probability review Lecture 1. Introduction and probability review 2 (1 1) 1. Introduction and probability review 1.1. What is Statistics?
More informationIntroduction to Bayesian Learning. Machine Learning Fall 2018
Introduction to Bayesian Learning Machine Learning Fall 2018 1 What we have seen so far What does it mean to learn? Mistake-driven learning Learning by counting (and bounding) number of mistakes PAC learnability
More informationMachine Learning CSE546 Sham Kakade University of Washington. Oct 4, What about continuous variables?
Linear Regression Machine Learning CSE546 Sham Kakade University of Washington Oct 4, 2016 1 What about continuous variables? Billionaire says: If I am measuring a continuous variable, what can you do
More informationBernoulli and Binomial Distributions. Notes. Bernoulli Trials. Bernoulli/Binomial Random Variables Bernoulli and Binomial Distributions.
Lecture 11 Text: A Course in Probability by Weiss 5.3 STAT 225 Introduction to Probability Models February 16, 2014 Whitney Huang Purdue University 11.1 Agenda 1 2 11.2 Bernoulli trials Many problems in
More informationLecture 7: Confidence interval and Normal approximation
Lecture 7: Confidence interval and Normal approximation 26th of November 2015 Confidence interval 26th of November 2015 1 / 23 Random sample and uncertainty Example: we aim at estimating the average height
More informationChapter 6: Large Random Samples Sections
Chapter 6: Large Random Samples Sections 6.1: Introduction 6.2: The Law of Large Numbers Skip p. 356-358 Skip p. 366-368 Skip 6.4: The correction for continuity Remember: The Midterm is October 25th in
More informationAnalysis of Engineering and Scientific Data. Semester
Analysis of Engineering and Scientific Data Semester 1 2019 Sabrina Streipert s.streipert@uq.edu.au Example: Draw a random number from the interval of real numbers [1, 3]. Let X represent the number. Each
More information(Practice Version) Midterm Exam 2
EECS 126 Probability and Random Processes University of California, Berkeley: Fall 2014 Kannan Ramchandran November 7, 2014 (Practice Version) Midterm Exam 2 Last name First name SID Rules. DO NOT open
More informationDiscrete Probability Distributions
Discrete Probability Distributions EGR 260 R. Van Til Industrial & Systems Engineering Dept. Copyright 2013. Robert P. Van Til. All rights reserved. 1 What s It All About? The behavior of many random processes
More informationLecture 2 Sep 5, 2017
CS 388R: Randomized Algorithms Fall 2017 Lecture 2 Sep 5, 2017 Prof. Eric Price Scribe: V. Orestis Papadigenopoulos and Patrick Rall NOTE: THESE NOTES HAVE NOT BEEN EDITED OR CHECKED FOR CORRECTNESS 1
More informationAdvanced Probability Theory (Math541)
Advanced Probability Theory (Math541) Instructor: Kani Chen (Classic)/Modern Probability Theory (1900-1960) Instructor: Kani Chen (HKUST) Advanced Probability Theory (Math541) 1 / 17 Primitive/Classic
More informationTopic 3: The Expectation of a Random Variable
Topic 3: The Expectation of a Random Variable Course 003, 2017 Page 0 Expectation of a discrete random variable Definition (Expectation of a discrete r.v.): The expected value (also called the expectation
More informationStats for Engineers: Lecture 4
Stats for Engineers: Lecture 4 Summary from last time Standard deviation σ measure spread of distribution μ Variance = (standard deviation) σ = var X = k μ P(X = k) k = k P X = k k μ σ σ k Discrete Random
More information6 The normal distribution, the central limit theorem and random samples
6 The normal distribution, the central limit theorem and random samples 6.1 The normal distribution We mentioned the normal (or Gaussian) distribution in Chapter 4. It has density f X (x) = 1 σ 1 2π e
More informationTwelfth Problem Assignment
EECS 401 Not Graded PROBLEM 1 Let X 1, X 2,... be a sequence of independent random variables that are uniformly distributed between 0 and 1. Consider a sequence defined by (a) Y n = max(x 1, X 2,..., X
More informationCSE 312 Final Review: Section AA
CSE 312 TAs December 8, 2011 General Information General Information Comprehensive Midterm General Information Comprehensive Midterm Heavily weighted toward material after the midterm Pre-Midterm Material
More informationMore discrete distributions
Will Monroe July 14, 217 with materials by Mehran Sahami and Chris Piech More discrete distributions Announcements: Problem Set 3 Posted yesterday on the course website. Due next Wednesday, 7/19, at 12:3pm
More informationSTA 291 Lecture 16. Normal distributions: ( mean and SD ) use table or web page. The sampling distribution of and are both (approximately) normal
STA 291 Lecture 16 Normal distributions: ( mean and SD ) use table or web page. The sampling distribution of and are both (approximately) normal X STA 291 - Lecture 16 1 Sampling Distributions Sampling
More informationLecture 14. Text: A Course in Probability by Weiss 5.6. STAT 225 Introduction to Probability Models February 23, Whitney Huang Purdue University
Lecture 14 Text: A Course in Probability by Weiss 5.6 STAT 225 Introduction to Probability Models February 23, 2014 Whitney Huang Purdue University 14.1 Agenda 14.2 Review So far, we have covered Bernoulli
More informationLecture 02: Combinatorics. Lisa Yan June 27, 2018
Lecture 02: Combinatorics Lisa Yan June 27, 2018 Announcements Sign up for Piazza https://piazza.com/stanford/summer2018/cs109 Sign up for Gradescope https://gradescope.com/courses/20179 (MWVBP7) Reference
More informationFundamentals. CS 281A: Statistical Learning Theory. Yangqing Jia. August, Based on tutorial slides by Lester Mackey and Ariel Kleiner
Fundamentals CS 281A: Statistical Learning Theory Yangqing Jia Based on tutorial slides by Lester Mackey and Ariel Kleiner August, 2011 Outline 1 Probability 2 Statistics 3 Linear Algebra 4 Optimization
More informationExpectation. DS GA 1002 Statistical and Mathematical Models. Carlos Fernandez-Granda
Expectation DS GA 1002 Statistical and Mathematical Models http://www.cims.nyu.edu/~cfgranda/pages/dsga1002_fall16 Carlos Fernandez-Granda Aim Describe random variables with a few numbers: mean, variance,
More informationGEOMETRIC -discrete A discrete random variable R counts number of times needed before an event occurs
STATISTICS 4 Summary Notes. Geometric and Exponential Distributions GEOMETRIC -discrete A discrete random variable R counts number of times needed before an event occurs P(X = x) = ( p) x p x =,, 3,...
More informationConvergence of Random Processes
Convergence of Random Processes DS GA 1002 Probability and Statistics for Data Science http://www.cims.nyu.edu/~cfgranda/pages/dsga1002_fall17 Carlos Fernandez-Granda Aim Define convergence for random
More informationTom Salisbury
MATH 2030 3.00MW Elementary Probability Course Notes Part V: Independence of Random Variables, Law of Large Numbers, Central Limit Theorem, Poisson distribution Geometric & Exponential distributions Tom
More informationConditional distributions
Conditional distributions Will Monroe July 6, 017 with materials by Mehran Sahami and Chris Piech Independence of discrete random variables Two random variables are independent if knowing the value of
More informationLecture 12. Poisson random variables
18.440: Lecture 12 Poisson random variables Scott Sheffield MIT 1 Outline Poisson random variable definition Poisson random variable properties Poisson random variable problems 2 Outline Poisson random
More informationLecture Tricks with Random Variables: The Law of Large Numbers & The Central Limit Theorem
Math 408 - Mathematical Statistics Lecture 9-10. Tricks with Random Variables: The Law of Large Numbers & The Central Limit Theorem February 6-8, 2013 Konstantin Zuev (USC) Math 408, Lecture 9-10 February
More informationQuestion Points Score Total: 137
Math 447 Test 1 SOLUTIONS Fall 2015 No books, no notes, only SOA-approved calculators. true/false or fill-in-the-blank question. You must show work, unless the question is a Name: Section: Question Points
More informationMachine Learning CSE546 Carlos Guestrin University of Washington. September 30, What about continuous variables?
Linear Regression Machine Learning CSE546 Carlos Guestrin University of Washington September 30, 2014 1 What about continuous variables? n Billionaire says: If I am measuring a continuous variable, what
More informationLecture 20 Random Samples 0/ 13
0/ 13 One of the most important concepts in statistics is that of a random sample. The definition of a random sample is rather abstract. However it is critical to understand the idea behind the definition,
More information2. Probability. Chris Piech and Mehran Sahami. Oct 2017
2. Probability Chris Piech and Mehran Sahami Oct 2017 1 Introduction It is that time in the quarter (it is still week one) when we get to talk about probability. Again we are going to build up from first
More informationSTAT Chapter 5 Continuous Distributions
STAT 270 - Chapter 5 Continuous Distributions June 27, 2012 Shirin Golchi () STAT270 June 27, 2012 1 / 59 Continuous rv s Definition: X is a continuous rv if it takes values in an interval, i.e., range
More informationStat 100a: Introduction to Probability. Outline for the day:
Stat 100a: Introduction to Probability. Outline for the day: 1. Midterm2 back. 2. CLT again. 3. Confidence intervals. 4. Sample size calculation. 5. Random walks. 6. Reflection principle. 7. Ballot theorem.
More informationMidterm Exam 2 (Solutions)
EECS 6 Probability and Random Processes University of California, Berkeley: Spring 07 Kannan Ramchandran March, 07 Midterm Exam (Solutions) Last name First name SID Name of student on your left: Name of
More informationTopic 7: Convergence of Random Variables
Topic 7: Convergence of Ranom Variables Course 003, 2016 Page 0 The Inference Problem So far, our starting point has been a given probability space (S, F, P). We now look at how to generate information
More informationCSE 312, 2011 Winter, W.L.Ruzzo. 6. random variables
CSE 312, 2011 Winter, W.L.Ruzzo 6. random variables random variables 23 numbered balls Ross 4.1 ex 1b 24 first head 25 probability mass functions 26 head count Let X be the number of heads observed in
More informationExpectation is linear. So far we saw that E(X + Y ) = E(X) + E(Y ). Let α R. Then,
Expectation is linear So far we saw that E(X + Y ) = E(X) + E(Y ). Let α R. Then, E(αX) = ω = ω (αx)(ω) Pr(ω) αx(ω) Pr(ω) = α ω X(ω) Pr(ω) = αe(x). Corollary. For α, β R, E(αX + βy ) = αe(x) + βe(y ).
More information