Machine Learning, Spring 2011: Homework 1 Solution

Size: px
Start display at page:

Download "Machine Learning, Spring 2011: Homework 1 Solution"

Transcription

1 Machie Learig, Sprig 011: Homework 1 Solutio February 1, 011 Istructios There are 3 questios o this assigmet. The last questio ivolves codig. Attach your code to the writeup. Please submit your homework as 3 separate sets of pages accordig to TAs, with your ame ad userid o each set. 1 Iformatio Gai, KL-divergece ad Etropy [Xi Che, 30 poits] 1. Whe we costruct a decisio tree, the ext attribute to split is the oe with maximum mutual iformatio (a.k.a. iformatio gai, which is defied i terms of etropy. I this problem, we will explore its coectio to KL-divergece. The KL-divergece from a distributio p(x to a distributio q(x ca be thought of as a distace measure from p to q: KL(p q = x p(x log q(x p(x. If p(x = q(x, the KL(p q = 0. Otherwise, KL(p q > 0. 1 We ca defie mutual iformatio as the KL-divergece from the observed joit distributio of X ad Y to the product of their margials: I(X, Y KL(p(x, y p(xp(y (a Show that this defiitio of mutual iformatio is equivalet to the oe give i class,. That is, show that I(X, Y = H(X H(X Y ad I(X, Y = H(Y H(Y X from the defiitio i terms of KL-divergece. From this defiitio, we ca easily see that mutual iformatio is symmetric, i.e. I(X, Y = I(Y, X. [10pt] (b Accordig to this defiitio, uder what coditios do we have that I(X, Y = 0. [5pt] (a KL(p q = x = x = y ( p(xp(y p(x, y log p(x, y y p(x, y(log p(x + log p(y log p(x, y y p(y log p(y + x p(x y p(y x log p(y x = H(Y H(Y X Equivalece to H(X H(X Y ca be show i a similar way. 1 For more details o KL-divergece, refer to Sectio 1.6 i Bishop. 1

2 (b Whe X ad Y are statistically idepedet, i.e. p(x, y = p(xp(y, I(X, Y = 0.. I the class, we defie the etropy based o a discrete radom variable X. Now cosider the case that X is a cotiuous radom variable with the probability desity fuctio p(x. The etropy is defied as: H(X = p(x l p(xdx Assume that X follows a Gaussia distributio with the mea µ ad variace σ, i.e. (a Please derive its etropy H(X. [10pt] p(x = 1 (x µ exp πσ σ (b Give a careful observatio for the etropy you derived ad please idicate oe property, which holds for the etropy for (ay discrete radom variable, but does ot hold here. [5pt] 0H(X = p(x l p(xdx = p(x ( 1 l(πσ = 1 (l(πσ + 1σ = 1 ( l(πσ + 1 (x µ σ p(x(x µ dx dx The last iequality is accordig to the variace of a stadard ormal distributio: σ = var(x = E ( (X µ = p(x(x µ dx Note that ulike the etropy for discrete variable which is always o-egative, whe σ < 1 πe, H(x < 0. Bayes Rule ad Poit Estimatio [Xi Che, 30 poits] 1. Assume the probability of a certai disease is The probability of testig positive give that a perso is ifected with the disease is 0.95 ad the probability of testig positive give the perso is ot ifected with the disease is (a Calculate the probability of testig positive. [5pt] (b Use Bayes Rule to calculate the probability of beig ifected with the disease give that the test is positive. [5pt] (a Give the iformatio i the problem, we have P (D = 0.01, P (T D = 0.95 ad P (T D = P (T = P (T D + P (T D = P (T DP (D + P (T DP (D = = 0.059

3 (b P (D T = P (T DP (D P (T = The Poisso distributio is a useful discrete distributio which ca be used to model the umber of occurreces of somethig per uit time. For example, i etworkig, the umber of packets to arrive i a give time widow is ofte assumed to follow a poisso distributio. If X is Poisso distributed, i.e. X P oisso(λ, its probability mass fuctio takes the followig form: P (X λ = λx e λ, X! It ca be show that if E(X = λ. Assume ow we have i.i.d. data poits from P oisso(λ: D = {X 1,..., X }. (For the purpose of this problem, you ca oly use the kowledge about the Poisso ad Gamma distributios provided i this problem. (a Show that the sample mea ˆλ = 1 X i is the maximum likelihood estimate (MLE of λ ad it is ubiased (E(ˆλ = λ. [8pt] (b Now let s be Bayesia ad put a prior distributio over λ. Assumig that λ follows a Gamma distributio with the parameters (α, β, its probability desity fuctio: p(λ α, β = βα Γ(α λα 1 e βλ, where Γ(α = (α 1! (here we assume α is a positive iteger. Compute the posterior distributio over λ. [6pt] (c Derive a aalytic expressio for the maximum a posterior (MAP of λ uder Gamma(α, β prior. [6pt] (a Write dow the log-likelihood (b l P (D λ = l e λ λ Xi X i! = λ + {X i l λ l(x i!}. The MLE ˆλ = arg max λ P (D λ = arg max λ l P (D λ, which ca be obtaied by settig the gradiet of l P (D λ with respect to λ to 0. More specifically: d dλ l P (D λ = + 1 X i = 0 = λ ˆλ = 1 X i Sice X 1,..., X are i.i.d. from P oisso(λ, for ay X i, E(X i = λ. ˆλ is ubiased because: ( 1 E(ˆλ = E X i = 1 E(X i = 1 λ = λ p(λ D P (D λp(λ ( = e λ λ Xi β α X i! Γ(α λα 1 e βλ λ Xi+α 1 e λ βλ Therefore, the posterior distributio p(λ D Gamma( X i + α, + β 3

4 (c The MAP λ = arg max λ p(λ D = arg max λ l p(λ D. Sice p(λ D λ Xi+α 1 e λ βλ, ( l p(λ D = X i + α 1 l λ ( + βλ + C, where C is a costat with respect to λ. Take the gradiet of l p(λ D with respect to λ ad set it to 0: d l p(λ D = X i + α 1 ( + β = 0 = λ = X i + α 1 dλ λ + β 3 Decisio Tree [Yi Zhag & Carl Doersch, 50 poits] I this questio, you will write our decisio tree code ad perform experimets with it. You will observe ad discuss the overfittig ad post-pruig of the decisio trees. Our data is a biary classificatio data set with discrete attributes, ad we oly require the decisio tree to be able to process this kid of data. All resources are provided i the file hw1 dt.zip, icludig the traiig, validatio (i.e., pruig ad testig data sets, a partially implemeted decisio tree codebase i C (with a few core parts removed, ad ecessary istructios to compile ad ru the codebase. Note: use of this codebase i ot required. If you are ot comfortable with codig i C, feel free to choose ay other laguage to implemet your ow decisio tree, as log as the tree ca perform the experimets we require o the particular data set we provided. While this codebase is ot a part of the questio, we icluded it so that, hopefully, may of you will be able to avoid the tedious implemetatio details ad focus istead o the iterestig parts of the decisio tree algorithm. Buildig the decisio tree: we build the decisio tree as we leared i the class. Give the traiig set, we will start from a sigle root ode with all the traiig examples assiged to it. The for each ode, if the assiged examples are ot pure (i.e., ot with the same label, we cosider further splitig this ode usig a best attribute. Selectig the best attribute for a give ode is the most importat part for buildig decisio trees, which is achieved by maximizig the iformatio gai of the split, or equivaletly miimizig the weighted average etropy after the splittig (i.e., the coditioal etropy give the attribute, show as H S (Y A i page 11 ad 1 of the slides. We will stop splitig a ode if: 1 the ode is pure; or we caot fid ay attribute that leads to a positive iformatio gai. Checkig a specific ode for pruig: Pruig a ode meas removig the subtree beeath it, keepig the ode as a leaf. As a result, all traiig examples assiged to the subtree are assiged to this ode. Examples assiged to the ode may ot all have the same label, ad i this case the label attached to this ode is the label of the majority class (ad examples of miority classes i this ode are misclassified, ad usually the classificatio accuracy o the traiig set will decrease. For performace reasos, we use a criterio differet from the lecture: give a specific ode ad the validatio set (i.e., the pruig set, we prue this ode if the classificatio accuracy of the resultig ew tree o the validatio set improves at least EP SILON. EP SILON is the threshold of miimal improvemet for pruig. For ow, we set EP SILON as (i.e., 0.5%, this default value has already bee set i our codebase. Post-pruig a decisio tree: top-dow ad bottom-up. I order to post-prue the etire decisio tree, we basically eed to perform a tree traversal, ad check all the odes alog the traversal. We cosider depth-first traversal, which ca be easily implemeted as oe fuctio via recursive calls. By placig the recursive calls at differet locatios of the fuctio, we ca make two choices: 1 check the curret ode before ivokig the recursive calls o its childre; ivoke the recursive calls o its childre before checkig the curret ode. Note that if a ode is checked ad actually prued, we will o loger travel to its childre. We iitially call the traversal fuctio at the root ode, ad clearly the two choices we metioed will lead to differet orders of checkig the tree odes. We call the first oe the top-dow approach sice it checks (ad tries to prue the paret ode before recursively checkig childre, ad we call the secod oe the bottom-up approach sice it checks childre before checkig the paret. Implemetatio ad the C codebase. The C codebase provides a decisio tree implemetatio (with a few parts removed by the TAs with much more fuctioality tha what we eed i this questio. So if you decide to use the C codebase, you oly eed to make chages o a few files (as detailed later without 4

5 really diggig ito every detail of this codebase. For a quick guide o how to compile ad ru the codebase, see quick start.txt i the hw1 dt.zip file. Data files. We use a oisy mushroom data set for this problem. Usig this data set, we will trai decisio trees to classify each mushroom as poisoous or ot, usig discrete features such as cap shape, cap color ad gill size. There are three data files i hw1 dt.zip: oisy10 trai.ssv, oisy10 valid.ssv, ad oisy10 test.ssv. They are traiig set, validatio set (i.e., pruig set, ad testig set. The format of each file is: first three lies are data statistics (umber of variables plus label, variable ames, properties of each variable, ad from the 4th lie is the data, where each lie is a example ad each colum is either the label (the first colum or a variable. You do t eed to worry about the data format if you use our codebase. 3.1 Complete the implemetatio [0 poits] To fully implemet the decisio tree usig the C codebase, there are maily two places i the codebase we eed to make chages: 1 etropy.c: the file implemetig ad usig the etropy fuctio to calculate iformatio gai ad choose the best splittig attribute whe buildig the decisio tree (search the commet YOU MUST MODIFY THIS FUNCTION i this file to fid the place to add your code ; pruedt.c: the file implemetig the post-pruig of the tree (search the commet YOU MUST MODIFY THIS FUNCTION i this file to fid the place to add your code. Prit the code you added i etropy.c ad prue-dt.c ad attach to your homework writeup. Note: if you choose to implemet your decisio tree without usig the codebase, just prit ad attach your code to the writeup. See the fuctio Etropy i etropy.c ad the fuctio PrueDecisioTree i prue-dt.c from the solutio code. Commo mistake 1: Not checkig the boudary coditio whe calculatig the etropy. If a ode cotais o positive example or o egative examples, we should directly retur 0.0 as the etropy istead of attemptig to calculate it, i.e., we do t wat to compute log (0. Commo mistake : Not covertig it to double before calculatig the quotiet of two umbers. C is ot as smart as Matlab ad R: we eed to make sure at least either umerator or deomiator is a floatig poit umber before computig their divisio. 3. Experimets with differet post-pruig strategies [0 poits] We ve discussed the top-dow ad the bottom-up approaches to travel ad prue the tree, which you should already implemeted i prue-dt.c. Ru the codebase (or your ow implemetatio with both approaches, usig the traiig set for buildig the tree, the validatio set for post-pruig, ad testig set to fially test the classificatio accuracy (agai, see quick start.txt for compilig ad ruig the codebase. Report i your homework 1 for the fully grow tree (without post-pruig: the tree size (i.e., the umber of odes ad the depth of the tree, the classificatio accuracy o the traiig set, ad the classificatio accuracy o the testig set; for the post-prued tree with top-dow approach: the tree size, the classificatio accuracy o the traiig set, ad the classificatio accuracy o the testig set; 3 for the post-prued tree with bottom-up approach: the tree size, the classificatio accuracy o the traiig set, ad the classificatio accuracy o the testig set. Note: all the iformatio ca be foud from the output whe we ru the codebase. [10 poits] Discuss how differet pruig approaches affect the size of the tree, traiig accuracy, ad testig accuracy. Also commet o the differece betwee the traiig accuracy ad the testig accuracy for each differet tree (i.e., the full tree ad two prued trees [10 poits]. 5

6 Table 1: Number of odes as EP SILON chages EP SILON = EP SILON = EP SILON = 0.01 EP SILON = 0.03 Top-Dow Bottom-Up The full-grow tree has 919 odes ad its depth is 1, with the traiig accuracy as 99.7% ad the testig accuracy as 79.6%. The top-dow prued tree has 116 odes ad its depth is 8, with the traiig accuracy as 89.6% ad the testig accuracy as 89.0%. The bottom-up prued tree has 681 odes ad its depth is 11, with the traiig accuracy as 9.% ad the testig accuracy as 88.1%. The top-dow pruig strategy tries to prue higher-level odes (i.e., those close to the root before attemptig to prue lower-level odes (i.e., those close to the leaves, so it is a more aggressive pruig strategy ad teds to produce smaller post-prued trees. Sice the resultig tree is small, i.e., a less complex model, the traiig accuracy will geerally be lower tha that of the full-grow tree (which overfits the traiig samples, but the testig accuracy will usually be higher tha that of the full-grow tree as the less complex model geeralizes to usee testig samples better. The bottom-up pruig strategy tries to prue childre odes before attemptig to prue parets, so it is ot as aggressive as the top-dow strategy ad thus will ted to prue less odes ad produce larger post-prued trees (compared to the top-dow prued trees. As a result, the traiig accuracy of the resultig tree will geerally be higher tha the top-dow prued tree (as it is larger ad more complex tha a top-dow prued tree, but lower tha that of a full-grow tree (sice the prued tree is still smaller ad thus less complex tha the full-grow tree. The testig accuracy of bottom-up prued tree is usually higher tha the full-grow tree (as pruig helps to prevet overfittig. It s difficult to predict which prued tree will have lower testig accuracy tha the other, because both of the followig cases could happe: (1 the top-dow prued tree is over-prued ad thus is too simple to get good testig accuracy; ( the bottom-up prued tree is ot sufficietly prued ad thus still overfit the traiig samples to certai degree. I our results, the bottom-up prued tree has slightly lower testig accuracy (88.1% tha that of the top-dow prued tree (89.0%, idicatig the case ( might happe here. The gap betwee the traiig accuracy ad the testig accuracy is a good idicator of how much the model overfits the traiig samples. As we ca see, the full grow tree with 919 odes has a large gap: 99.7% traiig accuracy ad 79.6% testig accuracy, idicatig serious overfittig. The bottom-up prued tree with 681 odes has a small gap: 9.% traiig accuracy ad 88.1% testig accuracy, idicatig slight overfittig. The top-dow prued tree with 116 odes has almost o gap: 89.6% traiig accuracy ad 89.0% testig accuracy, idicatig almost o overfittig. Fially, we wat to clarify that, although i this questio the smallest tree (i.e., the top-dow prued tree achieves the best testig accuracy, it is ot always the case that the simplest model is the best. Over-simplified model caot perform well. 3.3 Experimets with differet threshold EP SILON [10 poits] Whe checkig each ode, we require a miimal improvemet of validatio accuracy EP SILON for pruig. So far we use the default EP SILON = 0.005(i.e, 0.5%. For both top-dow ad bottom up pruig, chage EP SILON ad report the umber of odes i the prued tree for EP SILON = 0.001, 0.005, 0.01, Briefly explai your results (oe or two seteces will suffice. NOTE: i the codebase, EP SILON is defied i auxi.h: search #defie EPSILON to fid the locatio of EP SILON. See the Table 1 for detailed results. Geerally speakig, larger EP SILON will require more improvemet of validatio accuracy to approve a pruig, so icreasig EP SILON will ted to prue less odes ad produce larger post-prued trees. 6

1 Review of Probability & Statistics

1 Review of Probability & Statistics 1 Review of Probability & Statistics a. I a group of 000 people, it has bee reported that there are: 61 smokers 670 over 5 960 people who imbibe (drik alcohol) 86 smokers who imbibe 90 imbibers over 5

More information

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 5

CS434a/541a: Pattern Recognition Prof. Olga Veksler. Lecture 5 CS434a/54a: Patter Recogitio Prof. Olga Veksler Lecture 5 Today Itroductio to parameter estimatio Two methods for parameter estimatio Maimum Likelihood Estimatio Bayesia Estimatio Itroducto Bayesia Decisio

More information

10-701/ Machine Learning Mid-term Exam Solution

10-701/ Machine Learning Mid-term Exam Solution 0-70/5-78 Machie Learig Mid-term Exam Solutio Your Name: Your Adrew ID: True or False (Give oe setece explaatio) (20%). (F) For a cotiuous radom variable x ad its probability distributio fuctio p(x), it

More information

Machine Learning Brett Bernstein

Machine Learning Brett Bernstein Machie Learig Brett Berstei Week 2 Lecture: Cocept Check Exercises Starred problems are optioal. Excess Risk Decompositio 1. Let X = Y = {1, 2,..., 10}, A = {1,..., 10, 11} ad suppose the data distributio

More information

This exam contains 19 pages (including this cover page) and 10 questions. A Formulae sheet is provided with the exam.

This exam contains 19 pages (including this cover page) and 10 questions. A Formulae sheet is provided with the exam. Probability ad Statistics FS 07 Secod Sessio Exam 09.0.08 Time Limit: 80 Miutes Name: Studet ID: This exam cotais 9 pages (icludig this cover page) ad 0 questios. A Formulae sheet is provided with the

More information

FACULTY OF MATHEMATICAL STUDIES MATHEMATICS FOR PART I ENGINEERING. Lectures

FACULTY OF MATHEMATICAL STUDIES MATHEMATICS FOR PART I ENGINEERING. Lectures FACULTY OF MATHEMATICAL STUDIES MATHEMATICS FOR PART I ENGINEERING Lectures MODULE 5 STATISTICS II. Mea ad stadard error of sample data. Biomial distributio. Normal distributio 4. Samplig 5. Cofidece itervals

More information

EECS564 Estimation, Filtering, and Detection Hwk 2 Solns. Winter p θ (z) = (2θz + 1 θ), 0 z 1

EECS564 Estimation, Filtering, and Detection Hwk 2 Solns. Winter p θ (z) = (2θz + 1 θ), 0 z 1 EECS564 Estimatio, Filterig, ad Detectio Hwk 2 Sols. Witer 25 4. Let Z be a sigle observatio havig desity fuctio where. p (z) = (2z + ), z (a) Assumig that is a oradom parameter, fid ad plot the maximum

More information

Lecture 12: September 27

Lecture 12: September 27 36-705: Itermediate Statistics Fall 207 Lecturer: Siva Balakrisha Lecture 2: September 27 Today we will discuss sufficiecy i more detail ad the begi to discuss some geeral strategies for costructig estimators.

More information

Chapter 6 Principles of Data Reduction

Chapter 6 Principles of Data Reduction Chapter 6 for BST 695: Special Topics i Statistical Theory. Kui Zhag, 0 Chapter 6 Priciples of Data Reductio Sectio 6. Itroductio Goal: To summarize or reduce the data X, X,, X to get iformatio about a

More information

Goodness-of-Fit Tests and Categorical Data Analysis (Devore Chapter Fourteen)

Goodness-of-Fit Tests and Categorical Data Analysis (Devore Chapter Fourteen) Goodess-of-Fit Tests ad Categorical Data Aalysis (Devore Chapter Fourtee) MATH-252-01: Probability ad Statistics II Sprig 2019 Cotets 1 Chi-Squared Tests with Kow Probabilities 1 1.1 Chi-Squared Testig................

More information

Statistical Pattern Recognition

Statistical Pattern Recognition Statistical Patter Recogitio Classificatio: No-Parametric Modelig Hamid R. Rabiee Jafar Muhammadi Sprig 2014 http://ce.sharif.edu/courses/92-93/2/ce725-2/ Ageda Parametric Modelig No-Parametric Modelig

More information

ECE 8527: Introduction to Machine Learning and Pattern Recognition Midterm # 1. Vaishali Amin Fall, 2015

ECE 8527: Introduction to Machine Learning and Pattern Recognition Midterm # 1. Vaishali Amin Fall, 2015 ECE 8527: Itroductio to Machie Learig ad Patter Recogitio Midterm # 1 Vaishali Ami Fall, 2015 tue39624@temple.edu Problem No. 1: Cosider a two-class discrete distributio problem: ω 1 :{[0,0], [2,0], [2,2],

More information

7.1 Convergence of sequences of random variables

7.1 Convergence of sequences of random variables Chapter 7 Limit theorems Throughout this sectio we will assume a probability space (Ω, F, P), i which is defied a ifiite sequece of radom variables (X ) ad a radom variable X. The fact that for every ifiite

More information

Stat410 Probability and Statistics II (F16)

Stat410 Probability and Statistics II (F16) Some Basic Cocepts of Statistical Iferece (Sec 5.) Suppose we have a rv X that has a pdf/pmf deoted by f(x; θ) or p(x; θ), where θ is called the parameter. I previous lectures, we focus o probability problems

More information

CHAPTER 10 INFINITE SEQUENCES AND SERIES

CHAPTER 10 INFINITE SEQUENCES AND SERIES CHAPTER 10 INFINITE SEQUENCES AND SERIES 10.1 Sequeces 10.2 Ifiite Series 10.3 The Itegral Tests 10.4 Compariso Tests 10.5 The Ratio ad Root Tests 10.6 Alteratig Series: Absolute ad Coditioal Covergece

More information

6.3 Testing Series With Positive Terms

6.3 Testing Series With Positive Terms 6.3. TESTING SERIES WITH POSITIVE TERMS 307 6.3 Testig Series With Positive Terms 6.3. Review of what is kow up to ow I theory, testig a series a i for covergece amouts to fidig the i= sequece of partial

More information

Exponential Families and Bayesian Inference

Exponential Families and Bayesian Inference Computer Visio Expoetial Families ad Bayesia Iferece Lecture Expoetial Families A expoetial family of distributios is a d-parameter family f(x; havig the followig form: f(x; = h(xe g(t T (x B(, (. where

More information

Lecture 2: Monte Carlo Simulation

Lecture 2: Monte Carlo Simulation STAT/Q SCI 43: Itroductio to Resamplig ethods Sprig 27 Istructor: Ye-Chi Che Lecture 2: ote Carlo Simulatio 2 ote Carlo Itegratio Assume we wat to evaluate the followig itegratio: e x3 dx What ca we do?

More information

Hypothesis Testing. Evaluation of Performance of Learned h. Issues. Trade-off Between Bias and Variance

Hypothesis Testing. Evaluation of Performance of Learned h. Issues. Trade-off Between Bias and Variance Hypothesis Testig Empirically evaluatig accuracy of hypotheses: importat activity i ML. Three questios: Give observed accuracy over a sample set, how well does this estimate apply over additioal samples?

More information

Outline. CSCI-567: Machine Learning (Spring 2019) Outline. Prof. Victor Adamchik. Mar. 26, 2019

Outline. CSCI-567: Machine Learning (Spring 2019) Outline. Prof. Victor Adamchik. Mar. 26, 2019 Outlie CSCI-567: Machie Learig Sprig 209 Gaussia mixture models Prof. Victor Adamchik 2 Desity estimatio U of Souther Califoria Mar. 26, 209 3 Naive Bayes Revisited March 26, 209 / 57 March 26, 209 2 /

More information

Estimation for Complete Data

Estimation for Complete Data Estimatio for Complete Data complete data: there is o loss of iformatio durig study. complete idividual complete data= grouped data A complete idividual data is the oe i which the complete iformatio of

More information

Random Variables, Sampling and Estimation

Random Variables, Sampling and Estimation Chapter 1 Radom Variables, Samplig ad Estimatio 1.1 Itroductio This chapter will cover the most importat basic statistical theory you eed i order to uderstad the ecoometric material that will be comig

More information

STAT Homework 2 - Solutions

STAT Homework 2 - Solutions STAT-36700 Homework - Solutios Fall 08 September 4, 08 This cotais solutios for Homework. Please ote that we have icluded several additioal commets ad approaches to the problems to give you better isight.

More information

Solution of Final Exam : / Machine Learning

Solution of Final Exam : / Machine Learning Solutio of Fial Exam : 10-701/15-781 Machie Learig Fall 2004 Dec. 12th 2004 Your Adrew ID i capital letters: Your full ame: There are 9 questios. Some of them are easy ad some are more difficult. So, if

More information

Discrete Mathematics for CS Spring 2008 David Wagner Note 22

Discrete Mathematics for CS Spring 2008 David Wagner Note 22 CS 70 Discrete Mathematics for CS Sprig 2008 David Wager Note 22 I.I.D. Radom Variables Estimatig the bias of a coi Questio: We wat to estimate the proportio p of Democrats i the US populatio, by takig

More information

NYU Center for Data Science: DS-GA 1003 Machine Learning and Computational Statistics (Spring 2018)

NYU Center for Data Science: DS-GA 1003 Machine Learning and Computational Statistics (Spring 2018) NYU Ceter for Data Sciece: DS-GA 003 Machie Learig ad Computatioal Statistics (Sprig 208) Brett Berstei, David Roseberg, Be Jakubowski Jauary 20, 208 Istructios: Followig most lab ad lecture sectios, we

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2018 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

STAT Homework 1 - Solutions

STAT Homework 1 - Solutions STAT-36700 Homework 1 - Solutios Fall 018 September 11, 018 This cotais solutios for Homework 1. Please ote that we have icluded several additioal commets ad approaches to the problems to give you better

More information

MATH 472 / SPRING 2013 ASSIGNMENT 2: DUE FEBRUARY 4 FINALIZED

MATH 472 / SPRING 2013 ASSIGNMENT 2: DUE FEBRUARY 4 FINALIZED MATH 47 / SPRING 013 ASSIGNMENT : DUE FEBRUARY 4 FINALIZED Please iclude a cover sheet that provides a complete setece aswer to each the followig three questios: (a) I your opiio, what were the mai ideas

More information

Infinite Sequences and Series

Infinite Sequences and Series Chapter 6 Ifiite Sequeces ad Series 6.1 Ifiite Sequeces 6.1.1 Elemetary Cocepts Simply speakig, a sequece is a ordered list of umbers writte: {a 1, a 2, a 3,...a, a +1,...} where the elemets a i represet

More information

Math 10A final exam, December 16, 2016

Math 10A final exam, December 16, 2016 Please put away all books, calculators, cell phoes ad other devices. You may cosult a sigle two-sided sheet of otes. Please write carefully ad clearly, USING WORDS (ot just symbols). Remember that the

More information

Exercises Advanced Data Mining: Solutions

Exercises Advanced Data Mining: Solutions Exercises Advaced Data Miig: Solutios Exercise 1 Cosider the followig directed idepedece graph. 5 8 9 a) Give the factorizatio of P (X 1, X 2,..., X 9 ) correspodig to this idepedece graph. P (X) = 9 P

More information

If, for instance, we were required to test whether the population mean μ could be equal to a certain value μ

If, for instance, we were required to test whether the population mean μ could be equal to a certain value μ STATISTICAL INFERENCE INTRODUCTION Statistical iferece is that brach of Statistics i which oe typically makes a statemet about a populatio based upo the results of a sample. I oesample testig, we essetially

More information

Introduction to Machine Learning DIS10

Introduction to Machine Learning DIS10 CS 189 Fall 017 Itroductio to Machie Learig DIS10 1 Fu with Lagrage Multipliers (a) Miimize the fuctio such that f (x,y) = x + y x + y = 3. Solutio: The Lagragia is: L(x,y,λ) = x + y + λ(x + y 3) Takig

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2018 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Kinetics of Complex Reactions

Kinetics of Complex Reactions Kietics of Complex Reactios by Flick Colema Departmet of Chemistry Wellesley College Wellesley MA 28 wcolema@wellesley.edu Copyright Flick Colema 996. All rights reserved. You are welcome to use this documet

More information

7.1 Convergence of sequences of random variables

7.1 Convergence of sequences of random variables Chapter 7 Limit Theorems Throughout this sectio we will assume a probability space (, F, P), i which is defied a ifiite sequece of radom variables (X ) ad a radom variable X. The fact that for every ifiite

More information

15-780: Graduate Artificial Intelligence. Density estimation

15-780: Graduate Artificial Intelligence. Density estimation 5-780: Graduate Artificial Itelligece Desity estimatio Coditioal Probability Tables (CPT) But where do we get them? P(B)=.05 B P(E)=. E P(A B,E) )=.95 P(A B, E) =.85 P(A B,E) )=.5 P(A B, E) =.05 A P(J

More information

Mixtures of Gaussians and the EM Algorithm

Mixtures of Gaussians and the EM Algorithm Mixtures of Gaussias ad the EM Algorithm CSE 6363 Machie Learig Vassilis Athitsos Computer Sciece ad Egieerig Departmet Uiversity of Texas at Arligto 1 Gaussias A popular way to estimate probability desity

More information

Discrete Mathematics for CS Spring 2007 Luca Trevisan Lecture 22

Discrete Mathematics for CS Spring 2007 Luca Trevisan Lecture 22 CS 70 Discrete Mathematics for CS Sprig 2007 Luca Trevisa Lecture 22 Aother Importat Distributio The Geometric Distributio Questio: A biased coi with Heads probability p is tossed repeatedly util the first

More information

Let us give one more example of MLE. Example 3. The uniform distribution U[0, θ] on the interval [0, θ] has p.d.f.

Let us give one more example of MLE. Example 3. The uniform distribution U[0, θ] on the interval [0, θ] has p.d.f. Lecture 5 Let us give oe more example of MLE. Example 3. The uiform distributio U[0, ] o the iterval [0, ] has p.d.f. { 1 f(x =, 0 x, 0, otherwise The likelihood fuctio ϕ( = f(x i = 1 I(X 1,..., X [0,

More information

Statistical Inference (Chapter 10) Statistical inference = learn about a population based on the information provided by a sample.

Statistical Inference (Chapter 10) Statistical inference = learn about a population based on the information provided by a sample. Statistical Iferece (Chapter 10) Statistical iferece = lear about a populatio based o the iformatio provided by a sample. Populatio: The set of all values of a radom variable X of iterest. Characterized

More information

Information-based Feature Selection

Information-based Feature Selection Iformatio-based Feature Selectio Farza Faria, Abbas Kazeroui, Afshi Babveyh Email: {faria,abbask,afshib}@staford.edu 1 Itroductio Feature selectio is a topic of great iterest i applicatios dealig with

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2016 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Economics 241B Relation to Method of Moments and Maximum Likelihood OLSE as a Maximum Likelihood Estimator

Economics 241B Relation to Method of Moments and Maximum Likelihood OLSE as a Maximum Likelihood Estimator Ecoomics 24B Relatio to Method of Momets ad Maximum Likelihood OLSE as a Maximum Likelihood Estimator Uder Assumptio 5 we have speci ed the distributio of the error, so we ca estimate the model parameters

More information

Clustering. CM226: Machine Learning for Bioinformatics. Fall Sriram Sankararaman Acknowledgments: Fei Sha, Ameet Talwalkar.

Clustering. CM226: Machine Learning for Bioinformatics. Fall Sriram Sankararaman Acknowledgments: Fei Sha, Ameet Talwalkar. Clusterig CM226: Machie Learig for Bioiformatics. Fall 216 Sriram Sakararama Ackowledgmets: Fei Sha, Ameet Talwalkar Clusterig 1 / 42 Admiistratio HW 1 due o Moday. Email/post o CCLE if you have questios.

More information

MATH 320: Probability and Statistics 9. Estimation and Testing of Parameters. Readings: Pruim, Chapter 4

MATH 320: Probability and Statistics 9. Estimation and Testing of Parameters. Readings: Pruim, Chapter 4 MATH 30: Probability ad Statistics 9. Estimatio ad Testig of Parameters Estimatio ad Testig of Parameters We have bee dealig situatios i which we have full kowledge of the distributio of a radom variable.

More information

1 Inferential Methods for Correlation and Regression Analysis

1 Inferential Methods for Correlation and Regression Analysis 1 Iferetial Methods for Correlatio ad Regressio Aalysis I the chapter o Correlatio ad Regressio Aalysis tools for describig bivariate cotiuous data were itroduced. The sample Pearso Correlatio Coefficiet

More information

PSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 9

PSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 9 Hypothesis testig PSYCHOLOGICAL RESEARCH (PYC 34-C Lecture 9 Statistical iferece is that brach of Statistics i which oe typically makes a statemet about a populatio based upo the results of a sample. I

More information

Regression and generalization

Regression and generalization Regressio ad geeralizatio CE-717: Machie Learig Sharif Uiversity of Techology M. Soleymai Fall 2016 Curve fittig: probabilistic perspective Describig ucertaity over value of target variable as a probability

More information

Expectation-Maximization Algorithm.

Expectation-Maximization Algorithm. Expectatio-Maximizatio Algorithm. Petr Pošík Czech Techical Uiversity i Prague Faculty of Electrical Egieerig Dept. of Cyberetics MLE 2 Likelihood.........................................................................................................

More information

Problem Set 4 Due Oct, 12

Problem Set 4 Due Oct, 12 EE226: Radom Processes i Systems Lecturer: Jea C. Walrad Problem Set 4 Due Oct, 12 Fall 06 GSI: Assae Gueye This problem set essetially reviews detectio theory ad hypothesis testig ad some basic otios

More information

ECE 901 Lecture 14: Maximum Likelihood Estimation and Complexity Regularization

ECE 901 Lecture 14: Maximum Likelihood Estimation and Complexity Regularization ECE 90 Lecture 4: Maximum Likelihood Estimatio ad Complexity Regularizatio R Nowak 5/7/009 Review : Maximum Likelihood Estimatio We have iid observatios draw from a ukow distributio Y i iid p θ, i,, where

More information

Frequentist Inference

Frequentist Inference Frequetist Iferece The topics of the ext three sectios are useful applicatios of the Cetral Limit Theorem. Without kowig aythig about the uderlyig distributio of a sequece of radom variables {X i }, for

More information

Lecture 14: Graph Entropy

Lecture 14: Graph Entropy 15-859: Iformatio Theory ad Applicatios i TCS Sprig 2013 Lecture 14: Graph Etropy March 19, 2013 Lecturer: Mahdi Cheraghchi Scribe: Euiwoog Lee 1 Recap Bergma s boud o the permaet Shearer s Lemma Number

More information

CEE 522 Autumn Uncertainty Concepts for Geotechnical Engineering

CEE 522 Autumn Uncertainty Concepts for Geotechnical Engineering CEE 5 Autum 005 Ucertaity Cocepts for Geotechical Egieerig Basic Termiology Set A set is a collectio of (mutually exclusive) objects or evets. The sample space is the (collectively exhaustive) collectio

More information

AAEC/ECON 5126 FINAL EXAM: SOLUTIONS

AAEC/ECON 5126 FINAL EXAM: SOLUTIONS AAEC/ECON 5126 FINAL EXAM: SOLUTIONS SPRING 2015 / INSTRUCTOR: KLAUS MOELTNER This exam is ope-book, ope-otes, but please work strictly o your ow. Please make sure your ame is o every sheet you re hadig

More information

INF Introduction to classifiction Anne Solberg Based on Chapter 2 ( ) in Duda and Hart: Pattern Classification

INF Introduction to classifiction Anne Solberg Based on Chapter 2 ( ) in Duda and Hart: Pattern Classification INF 4300 90 Itroductio to classifictio Ae Solberg ae@ifiuioo Based o Chapter -6 i Duda ad Hart: atter Classificatio 90 INF 4300 Madator proect Mai task: classificatio You must implemet a classificatio

More information

First Year Quantitative Comp Exam Spring, Part I - 203A. f X (x) = 0 otherwise

First Year Quantitative Comp Exam Spring, Part I - 203A. f X (x) = 0 otherwise First Year Quatitative Comp Exam Sprig, 2012 Istructio: There are three parts. Aswer every questio i every part. Questio I-1 Part I - 203A A radom variable X is distributed with the margial desity: >

More information

Lecture 11 and 12: Basic estimation theory

Lecture 11 and 12: Basic estimation theory Lecture ad 2: Basic estimatio theory Sprig 202 - EE 94 Networked estimatio ad cotrol Prof. Kha March 2 202 I. MAXIMUM-LIKELIHOOD ESTIMATORS The maximum likelihood priciple is deceptively simple. Louis

More information

(A sequence also can be thought of as the list of function values attained for a function f :ℵ X, where f (n) = x n for n 1.) x 1 x N +k x N +4 x 3

(A sequence also can be thought of as the list of function values attained for a function f :ℵ X, where f (n) = x n for n 1.) x 1 x N +k x N +4 x 3 MATH 337 Sequeces Dr. Neal, WKU Let X be a metric space with distace fuctio d. We shall defie the geeral cocept of sequece ad limit i a metric space, the apply the results i particular to some special

More information

Lecture 19: Convergence

Lecture 19: Convergence Lecture 19: Covergece Asymptotic approach I statistical aalysis or iferece, a key to the success of fidig a good procedure is beig able to fid some momets ad/or distributios of various statistics. I may

More information

Lesson 10: Limits and Continuity

Lesson 10: Limits and Continuity www.scimsacademy.com Lesso 10: Limits ad Cotiuity SCIMS Academy 1 Limit of a fuctio The cocept of limit of a fuctio is cetral to all other cocepts i calculus (like cotiuity, derivative, defiite itegrals

More information

Econ 325/327 Notes on Sample Mean, Sample Proportion, Central Limit Theorem, Chi-square Distribution, Student s t distribution 1.

Econ 325/327 Notes on Sample Mean, Sample Proportion, Central Limit Theorem, Chi-square Distribution, Student s t distribution 1. Eco 325/327 Notes o Sample Mea, Sample Proportio, Cetral Limit Theorem, Chi-square Distributio, Studet s t distributio 1 Sample Mea By Hiro Kasahara We cosider a radom sample from a populatio. Defiitio

More information

1.010 Uncertainty in Engineering Fall 2008

1.010 Uncertainty in Engineering Fall 2008 MIT OpeCourseWare http://ocw.mit.edu.00 Ucertaity i Egieerig Fall 2008 For iformatio about citig these materials or our Terms of Use, visit: http://ocw.mit.edu.terms. .00 - Brief Notes # 9 Poit ad Iterval

More information

4.1 Data processing inequality

4.1 Data processing inequality ECE598: Iformatio-theoretic methods i high-dimesioal statistics Sprig 206 Lecture 4: Total variatio/iequalities betwee f-divergeces Lecturer: Yihog Wu Scribe: Matthew Tsao, Feb 8, 206 [Ed. Mar 22] Recall

More information

1 Approximating Integrals using Taylor Polynomials

1 Approximating Integrals using Taylor Polynomials Seughee Ye Ma 8: Week 7 Nov Week 7 Summary This week, we will lear how we ca approximate itegrals usig Taylor series ad umerical methods. Topics Page Approximatig Itegrals usig Taylor Polyomials. Defiitios................................................

More information

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.436J/15.085J Fall 2008 Lecture 19 11/17/2008 LAWS OF LARGE NUMBERS II THE STRONG LAW OF LARGE NUMBERS

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.436J/15.085J Fall 2008 Lecture 19 11/17/2008 LAWS OF LARGE NUMBERS II THE STRONG LAW OF LARGE NUMBERS MASSACHUSTTS INSTITUT OF TCHNOLOGY 6.436J/5.085J Fall 2008 Lecture 9 /7/2008 LAWS OF LARG NUMBRS II Cotets. The strog law of large umbers 2. The Cheroff boud TH STRONG LAW OF LARG NUMBRS While the weak

More information

Machine Learning Brett Bernstein

Machine Learning Brett Bernstein Machie Learig Brett Berstei Week Lecture: Cocept Check Exercises Starred problems are optioal. Statistical Learig Theory. Suppose A = Y = R ad X is some other set. Furthermore, assume P X Y is a discrete

More information

4. Partial Sums and the Central Limit Theorem

4. Partial Sums and the Central Limit Theorem 1 of 10 7/16/2009 6:05 AM Virtual Laboratories > 6. Radom Samples > 1 2 3 4 5 6 7 4. Partial Sums ad the Cetral Limit Theorem The cetral limit theorem ad the law of large umbers are the two fudametal theorems

More information

Lecture 10 October Minimaxity and least favorable prior sequences

Lecture 10 October Minimaxity and least favorable prior sequences STATS 300A: Theory of Statistics Fall 205 Lecture 0 October 22 Lecturer: Lester Mackey Scribe: Brya He, Rahul Makhijai Warig: These otes may cotai factual ad/or typographic errors. 0. Miimaxity ad least

More information

Lecture 4. Hw 1 and 2 will be reoped after class for every body. New deadline 4/20 Hw 3 and 4 online (Nima is lead)

Lecture 4. Hw 1 and 2 will be reoped after class for every body. New deadline 4/20 Hw 3 and 4 online (Nima is lead) Lecture 4 Homework Hw 1 ad 2 will be reoped after class for every body. New deadlie 4/20 Hw 3 ad 4 olie (Nima is lead) Pod-cast lecture o-lie Fial projects Nima will register groups ext week. Email/tell

More information

The Bayesian Learning Framework. Back to Maximum Likelihood. Naïve Bayes. Simple Example: Coin Tosses. Given a generative model

The Bayesian Learning Framework. Back to Maximum Likelihood. Naïve Bayes. Simple Example: Coin Tosses. Given a generative model Back to Maximum Likelihood Give a geerative model f (x, y = k) =π k f k (x) Usig a geerative modellig approach, we assume a parametric form for f k (x) =f (x; k ) ad compute the MLE θ of θ =(π k, k ) k=

More information

Lecture 7: Properties of Random Samples

Lecture 7: Properties of Random Samples Lecture 7: Properties of Radom Samples 1 Cotiued From Last Class Theorem 1.1. Let X 1, X,...X be a radom sample from a populatio with mea µ ad variace σ

More information

Classification Using Decision Trees. Jackknife Estimator: Example 1. Data Mining. Jackknife Estimator: Example 2(cont. Jackknife Estimator: Example 2

Classification Using Decision Trees. Jackknife Estimator: Example 1. Data Mining. Jackknife Estimator: Example 2(cont. Jackknife Estimator: Example 2 Data Miig CS 341, Sprig 2007 Lecture 8: Decisio tree algorithms Jackkife Estimator: Example 1 Estimate of mea for X={x 1, x 2, x 3,}, =3, g=3, m=1, θ = µ = (x( 1 + x 2 + x 3 )/3 θ 1 = (x( 2 + x 3 )/2,

More information

Lecture 11: Decision Trees

Lecture 11: Decision Trees ECE9 Sprig 7 Statistical Learig Theory Istructor: R. Nowak Lecture : Decisio Trees Miimum Complexity Pealized Fuctio Recall the basic results of the last lectures: let X ad Y deote the iput ad output spaces

More information

A sequence of numbers is a function whose domain is the positive integers. We can see that the sequence

A sequence of numbers is a function whose domain is the positive integers. We can see that the sequence Sequeces A sequece of umbers is a fuctio whose domai is the positive itegers. We ca see that the sequece,, 2, 2, 3, 3,... is a fuctio from the positive itegers whe we write the first sequece elemet as

More information

Probability and MLE.

Probability and MLE. 10-701 Probability ad MLE http://www.cs.cmu.edu/~pradeepr/701 (brief) itro to probability Basic otatios Radom variable - referrig to a elemet / evet whose status is ukow: A = it will rai tomorrow Domai

More information

SNAP Centre Workshop. Basic Algebraic Manipulation

SNAP Centre Workshop. Basic Algebraic Manipulation SNAP Cetre Workshop Basic Algebraic Maipulatio 8 Simplifyig Algebraic Expressios Whe a expressio is writte i the most compact maer possible, it is cosidered to be simplified. Not Simplified: x(x + 4x)

More information

The picture in figure 1.1 helps us to see that the area represents the distance traveled. Figure 1: Area represents distance travelled

The picture in figure 1.1 helps us to see that the area represents the distance traveled. Figure 1: Area represents distance travelled 1 Lecture : Area Area ad distace traveled Approximatig area by rectagles Summatio The area uder a parabola 1.1 Area ad distace Suppose we have the followig iformatio about the velocity of a particle, how

More information

Intro to Learning Theory

Intro to Learning Theory Lecture 1, October 18, 2016 Itro to Learig Theory Ruth Urer 1 Machie Learig ad Learig Theory Comig soo 2 Formal Framework 21 Basic otios I our formal model for machie learig, the istaces to be classified

More information

Expectation and Variance of a random variable

Expectation and Variance of a random variable Chapter 11 Expectatio ad Variace of a radom variable The aim of this lecture is to defie ad itroduce mathematical Expectatio ad variace of a fuctio of discrete & cotiuous radom variables ad the distributio

More information

Review Questions, Chapters 8, 9. f(y) = 0, elsewhere. F (y) = f Y(1) = n ( e y/θ) n 1 1 θ e y/θ = n θ e yn

Review Questions, Chapters 8, 9. f(y) = 0, elsewhere. F (y) = f Y(1) = n ( e y/θ) n 1 1 θ e y/θ = n θ e yn Stat 366 Lab 2 Solutios (September 2, 2006) page TA: Yury Petracheko, CAB 484, yuryp@ualberta.ca, http://www.ualberta.ca/ yuryp/ Review Questios, Chapters 8, 9 8.5 Suppose that Y, Y 2,..., Y deote a radom

More information

Ada Boost, Risk Bounds, Concentration Inequalities. 1 AdaBoost and Estimates of Conditional Probabilities

Ada Boost, Risk Bounds, Concentration Inequalities. 1 AdaBoost and Estimates of Conditional Probabilities CS8B/Stat4B Sprig 008) Statistical Learig Theory Lecture: Ada Boost, Risk Bouds, Cocetratio Iequalities Lecturer: Peter Bartlett Scribe: Subhrasu Maji AdaBoost ad Estimates of Coditioal Probabilities We

More information

Math 116 Second Midterm November 13, 2017

Math 116 Second Midterm November 13, 2017 Math 6 Secod Midterm November 3, 7 EXAM SOLUTIONS. Do ot ope this exam util you are told to do so.. Do ot write your ame aywhere o this exam. 3. This exam has pages icludig this cover. There are problems.

More information

Maximum Likelihood Estimation and Complexity Regularization

Maximum Likelihood Estimation and Complexity Regularization ECE90 Sprig 004 Statistical Regularizatio ad Learig Theory Lecture: 4 Maximum Likelihood Estimatio ad Complexity Regularizatio Lecturer: Rob Nowak Scribe: Pam Limpiti Review : Maximum Likelihood Estimatio

More information

4.3 Growth Rates of Solutions to Recurrences

4.3 Growth Rates of Solutions to Recurrences 4.3. GROWTH RATES OF SOLUTIONS TO RECURRENCES 81 4.3 Growth Rates of Solutios to Recurreces 4.3.1 Divide ad Coquer Algorithms Oe of the most basic ad powerful algorithmic techiques is divide ad coquer.

More information

Discrete probability distributions

Discrete probability distributions Discrete probability distributios I the chapter o probability we used the classical method to calculate the probability of various values of a radom variable. I some cases, however, we may be able to develop

More information

PRACTICE PROBLEMS FOR THE FINAL

PRACTICE PROBLEMS FOR THE FINAL PRACTICE PROBLEMS FOR THE FINAL Math 36Q Fall 25 Professor Hoh Below is a list of practice questios for the Fial Exam. I would suggest also goig over the practice problems ad exams for Exam ad Exam 2 to

More information

Introductory statistics

Introductory statistics CM9S: Machie Learig for Bioiformatics Lecture - 03/3/06 Itroductory statistics Lecturer: Sriram Sakararama Scribe: Sriram Sakararama We will provide a overview of statistical iferece focussig o the key

More information

ENGI 4421 Probability and Statistics Faculty of Engineering and Applied Science Problem Set 1 Solutions Descriptive Statistics. None at all!

ENGI 4421 Probability and Statistics Faculty of Engineering and Applied Science Problem Set 1 Solutions Descriptive Statistics. None at all! ENGI 44 Probability ad Statistics Faculty of Egieerig ad Applied Sciece Problem Set Solutios Descriptive Statistics. If, i the set of values {,, 3, 4, 5, 6, 7 } a error causes the value 5 to be replaced

More information

7-1. Chapter 4. Part I. Sampling Distributions and Confidence Intervals

7-1. Chapter 4. Part I. Sampling Distributions and Confidence Intervals 7-1 Chapter 4 Part I. Samplig Distributios ad Cofidece Itervals 1 7- Sectio 1. Samplig Distributio 7-3 Usig Statistics Statistical Iferece: Predict ad forecast values of populatio parameters... Test hypotheses

More information

Stat 421-SP2012 Interval Estimation Section

Stat 421-SP2012 Interval Estimation Section Stat 41-SP01 Iterval Estimatio Sectio 11.1-11. We ow uderstad (Chapter 10) how to fid poit estimators of a ukow parameter. o However, a poit estimate does ot provide ay iformatio about the ucertaity (possible

More information

Element sampling: Part 2

Element sampling: Part 2 Chapter 4 Elemet samplig: Part 2 4.1 Itroductio We ow cosider uequal probability samplig desigs which is very popular i practice. I the uequal probability samplig, we ca improve the efficiecy of the resultig

More information

There is no straightforward approach for choosing the warmup period l.

There is no straightforward approach for choosing the warmup period l. B. Maddah INDE 504 Discrete-Evet Simulatio Output Aalysis () Statistical Aalysis for Steady-State Parameters I a otermiatig simulatio, the iterest is i estimatig the log ru steady state measures of performace.

More information

ACCESS TO SCIENCE, ENGINEERING AND AGRICULTURE: MATHEMATICS 1 MATH00030 SEMESTER / Statistics

ACCESS TO SCIENCE, ENGINEERING AND AGRICULTURE: MATHEMATICS 1 MATH00030 SEMESTER / Statistics ACCESS TO SCIENCE, ENGINEERING AND AGRICULTURE: MATHEMATICS 1 MATH00030 SEMESTER 1 018/019 DR. ANTHONY BROWN 8. Statistics 8.1. Measures of Cetre: Mea, Media ad Mode. If we have a series of umbers the

More information

Grouping 2: Spectral and Agglomerative Clustering. CS 510 Lecture #16 April 2 nd, 2014

Grouping 2: Spectral and Agglomerative Clustering. CS 510 Lecture #16 April 2 nd, 2014 Groupig 2: Spectral ad Agglomerative Clusterig CS 510 Lecture #16 April 2 d, 2014 Groupig (review) Goal: Detect local image features (SIFT) Describe image patches aroud features SIFT, SURF, HoG, LBP, Group

More information

Outline. Linear regression. Regularization functions. Polynomial curve fitting. Stochastic gradient descent for regression. MLE for regression

Outline. Linear regression. Regularization functions. Polynomial curve fitting. Stochastic gradient descent for regression. MLE for regression REGRESSION 1 Outlie Liear regressio Regularizatio fuctios Polyomial curve fittig Stochastic gradiet descet for regressio MLE for regressio Step-wise forward regressio Regressio methods Statistical techiques

More information

Chapter 2 The Monte Carlo Method

Chapter 2 The Monte Carlo Method Chapter 2 The Mote Carlo Method The Mote Carlo Method stads for a broad class of computatioal algorithms that rely o radom sampligs. It is ofte used i physical ad mathematical problems ad is most useful

More information