Bayes, Oracle Bayes, and Empirical Bayes

Size: px
Start display at page:

Download "Bayes, Oracle Bayes, and Empirical Bayes"

Transcription

1 Bayes, Oracle Bayes, and Empirical Bayes

2 Empirical Bayes Inference Robbins (1951) Robbins (1956) Compound Decision Procedures Empirical Bayes Question How does empirical Bayes relate to Bayesian and frequentist inference? Intermediate framework: Oracle Bayes Bayes / Oracle Bayes / Empirical Bayes 2 / 37

3 A Familiar Empirical Bayes Setup Unknown prior g(θ) produces unseen parameters g(θ) θ 1, θ 2,..., θ N Each θ i independently gives observation x i N(θ i, 1), x = (x 1, x 2,..., x N ) Robbins Use all of x to estimate each θ i Bayes / Oracle Bayes / Empirical Bayes 3 / 37

4 Oracle Bayes Jiang and Zhang (2009) Oracle tells us the order statistic of the θ s, θ ord = ( ) θ (1), θ (2),..., θ (N) (but not which x i goes with which θ i ) We want to estimate ˆθ i that minimizes Expected Average Squared Error 1 N ( ) 2 EASE = E ˆθi θ i N Example: Two Towers Bayes / Oracle Bayes / Empirical Bayes 4 / 37 i=1

5 TWO TOWERS EXAMPLE: Oracle (Red), Observations (Black) g(theta) and fhat(x) theta and x Bayes / Oracle Bayes / Empirical Bayes 5 / 37

6 Using the Oracle ḡ(θ) empirical density of the θ s (probability 1 /N on θ (1), θ (2),..., θ (N) ) Bayes estimate (φ the standard normal density) ˆθḡ(x) = N / N θ (i) φ(x θ (i) ) φ(x θ (i) ) 1 1 Minimizes EASE among rules ˆθ(x) = t(x) Two Towers EASEḡ = MLE estimates ˆθ i = x i has EASE MLE = 1.0 Bayes / Oracle Bayes / Empirical Bayes 6 / 37

7 Empirical Bayes No Oracle Use x = (x 1, x 2,..., x N ) to form estimate ˆθ x (x) of Oracle rule ˆθḡ(x) Empirical Bayes regret EASE x EASEḡ = EBregret Two Towers EBregret = EASE x = = Bayes / Oracle Bayes / Empirical Bayes 7 / 37

8 The Frequentist Face of Empirical Bayes Oracle application is entirely frequentist: Minimizing EASE is a frequentist criterion Assumption θ i g(θ) is irrelevant! Question Is the 40% EASE reduction meaningful? Answer Yes for EASE, No for some other applications Bayes / Oracle Bayes / Empirical Bayes 8 / 37

9 Finite Bayes Inference θ i ind g(θ) and x i θ i ind N(θ i, 1) for i = 0, 1, 2,..., N We want posterior inference for θ 0 given x 0 and the sibling observations x = (x 1, x 2,..., x N ) If N then finite Bayes Bayes Next example: x 0 = 5, N = 50 Bayes / Oracle Bayes / Empirical Bayes 9 / 37

10 FINITE BAYES INFERENCE: 50 'sibling' observations to x0=5; what can we say about theta0? Frequency x0= x values Bayes / Oracle Bayes / Empirical Bayes 10 / 37

11 Red: likelihood N(x0=5,1); Green: Estimated prior from sibs, Black: Estimated posterior for theta0 posterior density prior x0=5 18% likelihood theta Bayes / Oracle Bayes / Empirical Bayes 11 / 37

12 Standard Bayes (Just one θ and x) Prior θ g(θ) Observe x p θ (x) [N(θ, 1)] Marginal density f(x) = X g(θ)p θ(x) dθ Bayesian inference x f(x) and θ x [ e g (x), v g (x) ] conditional expectation conditional variance For squared error: e g (x) is Bayes estimate ˆθ g (x) Bayes risk { (θ R g = E eg (x) ) } 2 = X v g(x)f(x) dx Rḡ = EASEḡ (= 0.563) Bayes / Oracle Bayes / Empirical Bayes 12 / 37

13 Tweedie s Formula (Efron, 2011) If x θ N(θ, 1): e g (x) = x + l (x) [l(x) = log f(x)] v g (x) = 1 + l (x) So ˆθ g (x) = x + l (x) and R g = [1 + l (x)] f(x) = 1 X [l (x)] 2 f(x) X Next: g(θ) = ḡ(θ), Two Towers Bayes / Oracle Bayes / Empirical Bayes 13 / 37

14 theta x: expectation e(x) + var(x)^.5, Two Towers example e(x)+ sqrt(v(x)) sqrt(v(x)) e(x) x value Bayes / Oracle Bayes / Empirical Bayes 14 / 37

15 Empirical Bayes Risk and Regret Idea Estimate ˆf(x) from data x = (x 1, x 2,..., x N ) ] ê(x) = x +ˆl (x) and ˆv(x) = 1 +ˆl (x) [ˆl(x) = logˆf(x) Empirical Bayes risk R(g, ê) = E { (ê(x) θ) 2} Lemma [ê(x) R(g, ê) = R g + eg (x) ] 2 f(x) dx X Bayes risk EBregret Only need f and its estimate ˆf Bayes / Oracle Bayes / Empirical Bayes 15 / 37

16 Estimating f(x): f-modeling Bin data x 1, x 2,..., x N y k = number of x s in bin k m k = midpoint of bin k ˆf = glm(y ns(m,df), poisson)$est/n Two Towers K = 44 bins (df = 5) Bayes / Oracle Bayes / Empirical Bayes 16 / 37

17 Histogram counts for the Two Towers data x[i], i=1:1500; K=44 bins ('bin[k]' in green) counts count y[k] midpoint m[k] x value Bayes / Oracle Bayes / Empirical Bayes 17 / 37

18 Fitted marginal density fhat(x) (1500) from glm(counts~ ns(x,5),poisson); green = true f(x) counts x value Bayes / Oracle Bayes / Empirical Bayes 18 / 37

19 Estimated conditional expectation e(x)=e{theta x} Two Towers, glm(y~ns(m,5),poisson); true e(x) in green EBregret=.007 ehat(x) estimated true x value Bayes / Oracle Bayes / Empirical Bayes 19 / 37

20 Data-based Formula for EBregret M = (ns(m,5), 1), kth row M k Ṁ, kth row dm k dm k ˆf = binned estimate of f(x) from glm(y M,poisson) Theorem E { EBregret } 1 { [M N trace diag (ˆf) ] 1 [Ṁ M diag (ˆf) } Ṁ] Decreases as 1 /N Doesn t account for bias of ê Bayes / Oracle Bayes / Empirical Bayes 20 / 37

21 The Poisson Case θ g( ) and x θ Poi(θ) [ pθ (x) = e θ θ x /x! ] f(x) = g(θ)p θ (x) dθ and θ x [ e g (x), v g (x) ] Robbins e g (x) = (x + 1)f(x + 1) f(x) and v g (x) = e g (x) [ e g (x + 1) e g (x) ] Bayes estimate ˆθg (x) = e g (x) with R g = f(x)v g (x) R [g, e(x)] = R g + f(x) [ e(x) e g (x) ] 2 X X Bayes / Oracle Bayes / Empirical Bayes 21 / 37

22 Butterfly Data Corbet in Malaysia, y 1 = 118 species trapped just one time each y 2 = 74 species trapped twice each... x y Bayes / Oracle Bayes / Empirical Bayes 22 / 37

23 Butterfly data: observed counts and fitted model glm(y~ns(x,5),poisson) red: zipf's law proportional 1/x 118 counts x value Bayes / Oracle Bayes / Empirical Bayes 23 / 37

24 Zipf s Law and Robbins Formula If f(x) = c/x then Robbins formula gives e g (x) = (x + 1)f(x + 1) f(x) = x So the MLE ˆθ i = x i is Bayes! Butterfly estimate R g = 5.24 and R(g, x) = 5.39 Bayes / Oracle Bayes / Empirical Bayes 24 / 37

25 Tweedie Poisson Estimation θ g( ) and x θ Poi(θ) f(x) = g(θ) ( e θ θ x /x! ) dθ λ = log θ is natural parameter E{λ x} = e g (x) = lgamma(x + 1) + l (x) Var{λ x} = v g (x) = lgamma(x + 1) + l (x) Same for truncated Poisson! Next: Butterfly data, ˆf from ns(df=5) Bayes / Oracle Bayes / Empirical Bayes 25 / 37

26 Tweedie for lambda: post expectation e(x) + v(x)^.5; Red line is (x,log(x)); Points from g modeling lambda Bayes risk =.363 MLE Risk.382 EB regret, x value Risk and Regret calcs for x in 1:20 Bayes / Oracle Bayes / Empirical Bayes 26 / 37

27 Empirical Bayes: g-modeling f-modeling: Never need to estimate prior g(θ) But only for criteria that depend just on f(x): EASE Tweedie Robbins false discovery rates... g-modeling (Efron, 2016): of prior g(θ) Direct parametric modeling Bayes / Oracle Bayes / Empirical Bayes 27 / 37

28 Parametric Models for g(θ) Exponential family log g(θ) = m(θ) β ( hidden GLM ) β a p-dimensional parameter; m(θ) say ns(θ,5) MLE ˆβ found by nonlinear maximization: β g β (θ) f β (x) y Mult(N, f β ) Trouble y Mult(N, f β ) not exponential family Advantage: Estimate g(θ x), Pr{θ > 2 x}, etc. Bayes / Oracle Bayes / Empirical Bayes 28 / 37

29 Estimated prior g(lambda) for the Butterfly data, lambda=log(theta); prior model M = ns(df=5) Bayes Risk & EBreg, lambda: theta: g(lambda) theta: lambda Bayes / Oracle Bayes / Empirical Bayes 29 / 37

30 Missing Species Problem Corbet How many new species in t more years of observation? Fisher, Good (Turing) Poisson process model f-modeling: E{#new} #old (Good and Toulmin, 1956) = Clever formula in terms of ˆf e θ 1 e θt/2 g(θ) dθ 1 e θ g-modeling: Substitute ĝ(θ) for g(θ) Bayes / Oracle Bayes / Empirical Bayes 30 / 37

31 Expected number new species seen in t years additional trapping; Red dots from Fisher Good Gaskins nonparametric formula # new species in two years additional years Bayes / Oracle Bayes / Empirical Bayes 31 / 37

32 A g-modeling Example gamnorm θ i Gamma 9 /3, x i N(θ i, 1), i = 1, 2,..., 3200 x = (x 1, x 2,..., x N ) for N = 15, 25, 50,..., 3200 g(θ) = ns(θ, 5) Next: posterior quantiles of θ 0 x 0 = 5 (finite Bayes inference as N increases) Bayes / Oracle Bayes / Empirical Bayes 32 / 37

33 POSTERIOR PERCENTILES of theta0 given x0=5 as the number of siblings increases; true gamnorm %iles at right limits log number siblings N Bayes / Oracle Bayes / Empirical Bayes 33 / 37

34 DTI Study 12 children, 6 dyslexic vs 6 normal controls N = brain voxels x i, x-value for ith voxel, x i N(θ i, 1) θ i = true effect size (= E{x i }) Bayes / Oracle Bayes / Empirical Bayes 34 / 37

35 DTI STUDY: x values for voxels, versus distance d from back of brain distance d from back of brain x value d=60,x=2.3 Bayes / Oracle Bayes / Empirical Bayes 35 / 37

36 g(theta0 x0=2.3) for decreasing sibling sets: (1) All (2) d in (3) (4) (5) only d=60 g(theta0 x0=2.3) x0= theta0 Bayes / Oracle Bayes / Empirical Bayes 36 / 37

37 References Efron, B. (2011). Tweedie s formula and selection bias. J. Amer. Statist. Assoc. 106: Efron, B. (2016). Empirical Bayes deconvolution estimates. Biometrika 103: Good, I. and Toulmin, G. (1956). The number of new species, and the increase in population coverage, when a sample is increased. Biometrika 43: Jiang, W. and Zhang, C.-H. (2009). General maximum likelihood empirical Bayes estimation of normal means. Ann. Statist. 37: Robbins, H. (1951). Asymptotically subminimax solutions of compound statistical decision problems. In Proceedings of the Second Berkeley Symposium on Mathematical Statistics and Probability, UC Press, Robbins, H. (1956). An empirical Bayes approach to statistics. In Proceedings of the Third Berkeley Symposium on Mathematical Statistics and Probability, , Vol. I. UC Press, Bayes / Oracle Bayes / Empirical Bayes 37 / 37

Minimum Mean-Square Error (MMSE) and Linear MMSE (LMMSE) Estimation

Minimum Mean-Square Error (MMSE) and Linear MMSE (LMMSE) Estimation Minimum Mean-Square Error (MMSE) and Linear MMSE (LMMSE) Estimation Outline: MMSE estimation, Linear MMSE (LMMSE) estimation, Geometric formulation of LMMSE estimation and orthogonality principle. Reading:

More information

By Lawrence D. Brown University of Pennsylvania

By Lawrence D. Brown University of Pennsylvania The Annals of Applied Statistics 2008, Vol. 2, No. 1, 113 152 DOI: 10.1214/07-AOAS138 c Institute of Mathematical Statistics, 2008 IN-SEASON PREDICTION OF BATTING AVERAGES: A FIELD TEST OF EMPIRICAL BAYES

More information

Remote Towers: Videopanorama Framerate Requirements Derived from Visual Discrimination of Deceleration During Simulated Aircraft Landing

Remote Towers: Videopanorama Framerate Requirements Derived from Visual Discrimination of Deceleration During Simulated Aircraft Landing www.dlr.de Chart 1 > SESARInno > Fürstenau RTOFramerate> 2012-11-30 Remote Towers: Videopanorama Framerate Requirements Derived from Visual Discrimination of Deceleration During Simulated Aircraft Landing

More information

Nonlife Actuarial Models. Chapter 7 Bühlmann Credibility

Nonlife Actuarial Models. Chapter 7 Bühlmann Credibility Nonlife Actuarial Models Chapter 7 Bühlmann Credibility Learning Objectives 1. Basic framework of Bühlmann credibility 2. Variance decomposition 3. Expected value of the process variance 4. Variance of

More information

Mathematics of Pari-Mutuel Wagering

Mathematics of Pari-Mutuel Wagering Millersville University of Pennsylvania April 17, 2014 Project Objectives Model the horse racing process to predict the outcome of a race. Use the win and exacta betting pools to estimate probabilities

More information

Hitters vs. Pitchers: A Comparison of Fantasy Baseball Player Performances Using Hierarchical Bayesian Models

Hitters vs. Pitchers: A Comparison of Fantasy Baseball Player Performances Using Hierarchical Bayesian Models Brigham Young University BYU ScholarsArchive All Theses and Dissertations 2012-04-17 Hitters vs. Pitchers: A Comparison of Fantasy Baseball Player Performances Using Hierarchical Bayesian Models Scott

More information

Chapter 20. Planning Accelerated Life Tests. William Q. Meeker and Luis A. Escobar Iowa State University and Louisiana State University

Chapter 20. Planning Accelerated Life Tests. William Q. Meeker and Luis A. Escobar Iowa State University and Louisiana State University Chapter 20 Planning Accelerated Life Tests William Q. Meeker and Luis A. Escobar Iowa State University and Louisiana State University Copyright 1998-2008 W. Q. Meeker and L. A. Escobar. Based on the authors

More information

SPATIAL STATISTICS A SPATIAL ANALYSIS AND COMPARISON OF NBA PLAYERS. Introduction

SPATIAL STATISTICS A SPATIAL ANALYSIS AND COMPARISON OF NBA PLAYERS. Introduction A SPATIAL ANALYSIS AND COMPARISON OF NBA PLAYERS KELLIN RUMSEY Introduction The 2016 National Basketball Association championship featured two of the leagues biggest names. The Golden State Warriors Stephen

More information

Example: sequence data set wit two loci [simula

Example: sequence data set wit two loci [simula Example: sequence data set wit two loci [simulated data] -- 1 Example: sequence data set wit two loci [simula MIGRATION RATE AND POPULATION SIZE ESTIMATION using the coalescent and maximum likelihood or

More information

Which On-Base Percentage Shows. the Highest True Ability of a. Baseball Player?

Which On-Base Percentage Shows. the Highest True Ability of a. Baseball Player? Which On-Base Percentage Shows the Highest True Ability of a Baseball Player? January 31, 2018 Abstract This paper looks at the true on-base ability of a baseball player given their on-base percentage.

More information

Data Set 7: Bioerosion by Parrotfish Background volume of bites The question:

Data Set 7: Bioerosion by Parrotfish Background volume of bites The question: Data Set 7: Bioerosion by Parrotfish Background Bioerosion of coral reefs results from animals taking bites out of the calcium-carbonate skeleton of the reef. Parrotfishes are major bioerosion agents,

More information

Journal of Quantitative Analysis in Sports

Journal of Quantitative Analysis in Sports Journal of Quantitative Analysis in Sports Volume 4, Issue 1 2008 Article 3 Streaky Hitting in Baseball Jim Albert Bowling Green State University, albert@bgnet.bgsu.edu Copyright c 2008 The Berkeley Electronic

More information

Unassisted breathing and death as competing events in critical care trials

Unassisted breathing and death as competing events in critical care trials Unassisted breathing and death as competing events in critical care trials William Checkley, MD, PhD Johns Hopkins University November 22, 2011 wcheckl1@jhmi.edu Objectives Jointly model the frequency

More information

knn & Naïve Bayes Hongning Wang

knn & Naïve Bayes Hongning Wang knn & Naïve Bayes Hongning Wang CS@UVa Today s lecture Instance-based classifiers k nearest neighbors Non-parametric learning algorithm Model-based classifiers Naïve Bayes classifier A generative model

More information

James-Stein Estimation in Baseball

James-Stein Estimation in Baseball The University of Akron IdeaExchange@UAkron Honors Research Projects The Dr. Gary B. and Pamela S. Williams Honors College Spring 2018 James-Stein Estimation in Baseball Isaac Matejin iam15@zips.uakron.edu

More information

CS145: INTRODUCTION TO DATA MINING

CS145: INTRODUCTION TO DATA MINING CS145: INTRODUCTION TO DATA MINING 3: Vector Data: Logistic Regression Instructor: Yizhou Sun yzsun@cs.ucla.edu October 9, 2017 Methods to Learn Vector Data Set Data Sequence Data Text Data Classification

More information

Safety Impacts of Road Diets in Iowa

Safety Impacts of Road Diets in Iowa University of Kentucky From the SelectedWorks of Reginald R. Souleyrette December, 2006 Safety Impacts of Road Diets in Iowa Thomas B. Stout, Iowa State University Michael D. Pawlovich Reginald R. Souleyrette,

More information

Estimating the Probability of Winning an NFL Game Using Random Forests

Estimating the Probability of Winning an NFL Game Using Random Forests Estimating the Probability of Winning an NFL Game Using Random Forests Dale Zimmerman February 17, 2017 2 Brian Burke s NFL win probability metric May be found at www.advancednflstats.com, but the site

More information

A.6 Graphic Files. A Scale-Free Transportation Network Explains the City-Size Distribution. Figure 6. The Interstate route map (abridged).

A.6 Graphic Files. A Scale-Free Transportation Network Explains the City-Size Distribution. Figure 6. The Interstate route map (abridged). A Scale-Free Transportation Network Explains the City-Size Distribution A. Graphic Files Figure. The Interstate route map (abridged). Figure. A typical airline s route configuration (pre-merger Continental

More information

The "Fair" Triathlon: Equating Standard Deviations Using Bayesian Nonlinear Models

The Fair Triathlon: Equating Standard Deviations Using Bayesian Nonlinear Models Brigham Young University BYU ScholarsArchive All Faculty Publications 2006-01-01 The "Fair" Triathlon: Equating Standard Deviations Using Bayesian Nonlinear Models Gilbert W. Fellingham gwf@byu.edu C.

More information

ACCIDENT MODIFICATION FACTORS FOR MEDIANS ON FREEWAYS AND MULTILANE RURAL HIGHWAYS IN TEXAS

ACCIDENT MODIFICATION FACTORS FOR MEDIANS ON FREEWAYS AND MULTILANE RURAL HIGHWAYS IN TEXAS Fitzpatrick, Lord, Park ACCIDENT MODIFICATION FACTORS FOR MEDIANS ON FREEWAYS AND MULTILANE RURAL HIGHWAYS IN TEXAS Kay Fitzpatrick Senior Research Engineer Texas Transportation Institute, 335 TAMU College

More information

EVALUATING THE SAFETY EFFECTS OF DAYLIGHT SAVINGS TIME ON FATAL AND NON-FATAL INJURY CRASHES IN TEXAS

EVALUATING THE SAFETY EFFECTS OF DAYLIGHT SAVINGS TIME ON FATAL AND NON-FATAL INJURY CRASHES IN TEXAS EVALUATING THE SAFETY EFFECTS OF DAYLIGHT SAVINGS TIME ON FATAL AND NON-FATAL INJURY CRASHES IN TEXAS by Charles R. Stevens Jr. 1 Texas Transportation Institute Texas A&M University System 701 N. Post

More information

Optimizing Cyclist Parking in a Closed System

Optimizing Cyclist Parking in a Closed System Optimizing Cyclist Parking in a Closed System Letu Qingge, Killian Smith Gianforte School of Computing, Montana State University, Bozeman, MT 59717, USA Abstract. In this paper, we consider the two different

More information

United States Commercial Vertical Line Vessel Standardized Catch Rates of Red Grouper in the US South Atlantic,

United States Commercial Vertical Line Vessel Standardized Catch Rates of Red Grouper in the US South Atlantic, SEDAR19-DW-14 United States Commercial Vertical Line Vessel Standardized Catch Rates of Red Grouper in the US South Atlantic, 1993-2008 Kevin McCarthy and Neil Baertlein National Marine Fisheries Service,

More information

Naïve Bayes. Robot Image Credit: Viktoriya Sukhanova 123RF.com

Naïve Bayes. Robot Image Credit: Viktoriya Sukhanova 123RF.com Naïve Bayes These slides were assembled by Byron Boots, with only minor modifications from Eric Eaton s slides and grateful acknowledgement to the many others who made their course materials freely available

More information

AN INDEX OF ABUNDANCE OF BLUEFIN TUNA IN THE NORTHWEST ATLANTIC OCEAN FROM COMBINED CANADA-U.S. PELAGIC LONGLINE DATA

AN INDEX OF ABUNDANCE OF BLUEFIN TUNA IN THE NORTHWEST ATLANTIC OCEAN FROM COMBINED CANADA-U.S. PELAGIC LONGLINE DATA SCRS/2015/171 AN INDEX OF ABUNDANCE OF BLUEFIN TUNA IN THE NORTHWEST ATLANTIC OCEAN FROM COMBINED CANADA-U.S. PELAGIC LONGLINE DATA Matthew Lauretta 1, Alex Hanke 2, and Irene Andrushchenko 2 SUMMARY During

More information

arxiv:math/ v4 [math.st] 7 Mar 2006

arxiv:math/ v4 [math.st] 7 Mar 2006 A DERIVATION OF THE PYTHAGOREAN WON-LOSS FORMULA IN BASEBALL arxiv:math/009698v4 [math.st] 7 Mar 06 STEVEN J. MILLER Abstract. It has been noted that in many professional sports leagues a good predictor

More information

Lecture 5. Optimisation. Regularisation

Lecture 5. Optimisation. Regularisation Lecture 5. Optimisation. Regularisation COMP90051 Statistical Machine Learning Semester 2, 2017 Lecturer: Andrey Kan Copyright: University of Melbourne Iterative optimisation Loss functions Coordinate

More information

CPUE standardization of black marlin (Makaira indica) caught by Taiwanese large scale longline fishery in the Indian Ocean

CPUE standardization of black marlin (Makaira indica) caught by Taiwanese large scale longline fishery in the Indian Ocean CPUE standardization of black marlin (Makaira indica) caught by Taiwanese large scale longline fishery in the Indian Ocean Sheng-Ping Wang Department of Environmental Biology and Fisheries Science, National

More information

Confidence Interval Notes Calculating Confidence Intervals

Confidence Interval Notes Calculating Confidence Intervals Confidence Interval Notes Calculating Confidence Intervals Calculating One-Population Mean Confidence Intervals for Quantitative Data It is always best to use a computer program to make these calculations,

More information

Addendum to SEDAR16-DW-22

Addendum to SEDAR16-DW-22 Addendum to SEDAR16-DW-22 Introduction Six king mackerel indices of abundance, two for each region Gulf of Mexico, South Atlantic, and Mixing Zone, were constructed for the SEDAR16 data workshop using

More information

STAT/MATH 395 PROBABILITY II

STAT/MATH 395 PROBABILITY II STAT/MATH 395 PROBABILITY II Quick review on Discrete Random Variables Néhémy Lim University of Washington Winter 2017 Example Pick 5 toppings from a total of 15. Give the sample space Ω of the experiment

More information

Announcements. Lecture 19: Inference for SLR & Transformations. Online quiz 7 - commonly missed questions

Announcements. Lecture 19: Inference for SLR & Transformations. Online quiz 7 - commonly missed questions Announcements Announcements Lecture 19: Inference for SLR & Statistics 101 Mine Çetinkaya-Rundel April 3, 2012 HW 7 due Thursday. Correlation guessing game - ends on April 12 at noon. Winner will be announced

More information

Evaluation of the Safety Effectiveness of the Conversion of Two-Lane Roadways to Four-Lane Divided Roadways

Evaluation of the Safety Effectiveness of the Conversion of Two-Lane Roadways to Four-Lane Divided Roadways Evaluation of the Safety Effectiveness of the Conversion of Two-Lane Roadways to Four-Lane Divided Roadways Bayesian Versus Empirical Bayes Mohamed M. Ahmed, Mohamed Abdel-Aty, and Juneyoung Park This

More information

Bayesian Methods: Naïve Bayes

Bayesian Methods: Naïve Bayes Bayesian Methods: Naïve Bayes Nicholas Ruozzi University of Texas at Dallas based on the slides of Vibhav Gogate Last Time Parameter learning Learning the parameter of a simple coin flipping model Prior

More information

Naïve Bayes. Robot Image Credit: Viktoriya Sukhanova 123RF.com

Naïve Bayes. Robot Image Credit: Viktoriya Sukhanova 123RF.com Naïve Bayes These slides were assembled by Eric Eaton, with grateful acknowledgement of the many others who made their course materials freely available online. Feel free to reuse or adapt these slides

More information

Standing Between a Bayesian and a Frequentist: An Emperical Bayes Exploration of Movies, Baseball, and Long Beach Basketball.

Standing Between a Bayesian and a Frequentist: An Emperical Bayes Exploration of Movies, Baseball, and Long Beach Basketball. Standing Between a Bayesian and a Frequentist: An Emperical Bayes Exploration of Movies, Baseball, and Long Beach Basketball Arthur Berg Pennsylvania State University Arthur Berg Standing Between a Bayesian

More information

A definition of depth for functional observations

A definition of depth for functional observations A definition of depth for functional observations Sara López-Pintado and Juan Romo Departamento de Estadística y Econometría Universidad Carlos III de Madrid Madrid, Spain A definition of depth for functional

More information

Lab 5: Descriptive Statistics

Lab 5: Descriptive Statistics Page 1 Technical Math II Lab 5: Descriptive Stats Lab 5: Descriptive Statistics Purpose: To gain experience in the descriptive statistical analysis of a large (173 scores) data set. You should do most

More information

A few things to remember about ANOVA

A few things to remember about ANOVA A few things to remember about ANOVA 1) The F-test that is performed is always 1-tailed. This is because your alternative hypothesis is always that the between group variation is greater than the within

More information

Safety Effectiveness of Pedestrian Crossing Treatments

Safety Effectiveness of Pedestrian Crossing Treatments Portland State University PDXScholar TREC Friday Seminar Series Transportation Research and Education Center (TREC) 10-13-2017 Safety Effectiveness of Pedestrian Crossing Treatments Christopher Monsere

More information

Modelling Exposure at Default Without Conversion Factors for Revolving Facilities

Modelling Exposure at Default Without Conversion Factors for Revolving Facilities Modelling Exposure at Default Without Conversion Factors for Revolving Facilities Mark Thackham Credit Scoring and Credit Control XV, Edinburgh, August 2017 1 / 27 Objective The objective of this presentation

More information

c 2016 Arash Khatibi

c 2016 Arash Khatibi c 2016 Arash Khatibi MODELING THE WINNING SEED DISTRIBUTION OF THE NCAA BASKETBALL TOURNAMENT BY ARASH KHATIBI THESIS Submitted in partial fulfillment of the requirements for the degree of Master of Science

More information

On the convergence of fitting algorithms in computer vision

On the convergence of fitting algorithms in computer vision On the convergence of fitting algorithms in computer vision N. Chernov Department of Mathematics University of Alabama at Birmingham Birmingham, AL 35294 chernov@math.uab.edu Abstract We investigate several

More information

Useful Tools in Mosquito Surveillance

Useful Tools in Mosquito Surveillance Useful Tools in Mosquito Surveillance Denominators Infection Rates (MIR and MLE s) Data Smoothing Vector Index Denominator The denominator is the lower portion of a fraction used to calculate a rate or

More information

STT 315 Section /19/2014

STT 315 Section /19/2014 Name: PID: A STT 315 Section 101 05/19/2014 Quiz 1A 50 minutes 1. A survey by an electric company contains questions on the following: Age of household head, Gender of household head and use of electric

More information

Standardized catch rates of U.S. blueline tilefish (Caulolatilus microps) from commercial logbook longline data

Standardized catch rates of U.S. blueline tilefish (Caulolatilus microps) from commercial logbook longline data Standardized catch rates of U.S. blueline tilefish (Caulolatilus microps) from commercial logbook longline data Sustainable Fisheries Branch, National Marine Fisheries Service, Southeast Fisheries Science

More information

Elizabeth A. Babcock University of Miami, Rosenstiel School of Marine and Atmospheric Sciences, 4600 Rickenbacker Cswy. Miami, FL USA

Elizabeth A. Babcock University of Miami, Rosenstiel School of Marine and Atmospheric Sciences, 4600 Rickenbacker Cswy. Miami, FL USA TOWARD CATCH QUOTAS FOR SPINY LOBSTER (Panulirus argus) AT GLOVER S REEF MARINE RESERVE Elizabeth A. Babcock University of Miami, Rosenstiel School of Marine and Atmospheric Sciences, 4600 Rickenbacker

More information

An Application of Signal Detection Theory for Understanding Driver Behavior at Highway-Rail Grade Crossings

An Application of Signal Detection Theory for Understanding Driver Behavior at Highway-Rail Grade Crossings An Application of Signal Detection Theory for Understanding Driver Behavior at Highway-Rail Grade Crossings Michelle Yeh and Jordan Multer United States Department of Transportation Volpe National Transportation

More information

Hitting with Runners in Scoring Position

Hitting with Runners in Scoring Position Hitting with Runners in Scoring Position Jim Albert Department of Mathematics and Statistics Bowling Green State University November 25, 2001 Abstract Sportscasters typically tell us about the batting

More information

Physics 122 Class #6 Outline

Physics 122 Class #6 Outline Physics 122 Class #6 Outline Announcements/Reading Assignment Speed of waves in a string Wave energy and intensity History and snapshot plots Properties of sinusoidal waves Velocity of the string (not

More information

Systematic Review and Meta-analysis of Bicycle Helmet Efficacy to Mitigate Head, Face and Neck Injuries

Systematic Review and Meta-analysis of Bicycle Helmet Efficacy to Mitigate Head, Face and Neck Injuries Systematic Review and Meta-analysis of Bicycle Helmet Efficacy to Mitigate Head, Face and Neck Injuries Prudence Creighton & Jake Olivier MATHEMATICS & THE UNIVERSITY OF NEW STATISTICS SOUTH WALES Creighton

More information

Estimating Paratransit Demand Forecasting Models Using ACS Disability and Income Data

Estimating Paratransit Demand Forecasting Models Using ACS Disability and Income Data Estimating Paratransit Demand Forecasting Models Using ACS Disability and Income Data Presenter: Daniel Rodríguez Román University of Puerto Rico, Mayagüez Co-author: Sarah V. Hernandez University of Arkansas,

More information

arxiv: v1 [stat.ap] 30 May 2017 Summary of Analysis

arxiv: v1 [stat.ap] 30 May 2017 Summary of Analysis A Hierarchical Model to Evaluate Policies for Reducing Vehicle Speed in Major American Cities Jonathan Auerbach, Christopher Eshleman and Rob Trangucci May 30, 2017 arxiv:1705.10876v1 [stat.ap] 30 May

More information

Dynamic Bradley Terry modelling of sports tournaments

Dynamic Bradley Terry modelling of sports tournaments Appl. Statist. (2013) 62, Part 1, pp. 135 150 Dynamic Bradley Terry modelling of sports tournaments Manuela Cattelan, Università di Padova, Italy Cristiano Varin Università Ca Foscari, Venice, Italy and

More information

Bayesian model averaging with change points to assess the impact of vaccination and public health interventions

Bayesian model averaging with change points to assess the impact of vaccination and public health interventions Bayesian model averaging with change points to assess the impact of vaccination and public health interventions SUPPLEMENTARY METHODS Data sources U.S. hospitalization data were obtained from the Healthcare

More information

Financial Econometrics and Volatility Models

Financial Econometrics and Volatility Models Financial Econometrics and Volatility Models Ei Eric Zivot Updated: May 3, 2010 Eric Zivot. All Rights Reserved. Example Data: S&P 500 returns Daily Closing Prices 100 150 200 250 300 1960 1965 1970 1975

More information

By Shane T. Jensen, Kenneth E. Shirley and Abraham J. Wyner. University of Pennsylvania

By Shane T. Jensen, Kenneth E. Shirley and Abraham J. Wyner. University of Pennsylvania The Annals of Applied Statistics 2009, Vol. 3, No. 2, 491 520 DOI: 10.1214/08-AOAS228 c Institute of Mathematical Statistics, 2009 BAYESBALL: A BAYESIAN HIERARCHICAL MODEL FOR EVALUATING FIELDING IN MAJOR

More information

Safety Effects of Left-Turn Phasing Schemes at High-Speed Intersections

Safety Effects of Left-Turn Phasing Schemes at High-Speed Intersections 2007-03 Safety Effects of Left-Turn Phasing Schemes at High-Speed Intersections Take the steps... Research...Knowledge...Innovative Solutions! Transportation Research Technical Report Documentation Page

More information

Report on the Comparative Fishing Trial Between the Gadus Atlantica and Teleost

Report on the Comparative Fishing Trial Between the Gadus Atlantica and Teleost NAFO Sci. Coun. Studies, 29: 81 92 Report on the Comparative Fishing Trial Between the Gadus Atlantica and Teleost William G. Warren Science Branch, Department of Fisheries and Oceans P. O. Box 5667, St.

More information

Logistic Regression. Hongning Wang

Logistic Regression. Hongning Wang Logistic Regression Hongning Wang CS@UVa Today s lecture Logistic regression model A discriminative classification model Two different perspectives to derive the model Parameter estimation CS@UVa CS 6501:

More information

USING DELTA-GAMMA GENERALIZED LINEAR MODELS TO STANDARDIZE CATCH RATES OF YELLOWFIN TUNA CAUGHT BY BRAZILIAN BAIT-BOATS

USING DELTA-GAMMA GENERALIZED LINEAR MODELS TO STANDARDIZE CATCH RATES OF YELLOWFIN TUNA CAUGHT BY BRAZILIAN BAIT-BOATS SCRS/2008/166 USING DELTA-GAMMA GENERALIZED LINEAR MODELS TO STANDARDIZE CATCH RATES OF YELLOWFIN TUNA CAUGHT BY BRAZILIAN BAIT-BOATS Humber A. Andrade 1 SUMMARY In order to standardize catch per unit

More information

SAFETY IMPACTS OF THE ACTUATED SIGNAL CONTROL AT URBAN INTERSECTIONS

SAFETY IMPACTS OF THE ACTUATED SIGNAL CONTROL AT URBAN INTERSECTIONS SANG HYUK LEE, Ph.D. E-mail: sanghyuklee862@gmail.com DAESEOK HAN, Ph.D. E-mail: handaeseok@kict.re.kr Korea Institute of Civil Engineering and Building Technology 283, Goyangdae-Ro, Ilsanseo-Gu, Goyang-Si,

More information

Introduction to Analysis of Variance (ANOVA) The Structural Model, The Summary Table, and the One- Way ANOVA

Introduction to Analysis of Variance (ANOVA) The Structural Model, The Summary Table, and the One- Way ANOVA Introduction to Analysis of Variance (ANOVA) The Structural Model, The Summary Table, and the One- Way ANOVA Limitations of the t-test Although the t-test is commonly used, it has limitations Can only

More information

Bridge Decomposition of Restriction Measures Tom Alberts joint work with Hugo Duminil (ENS) with lots of help from Wendelin Werner University of

Bridge Decomposition of Restriction Measures Tom Alberts joint work with Hugo Duminil (ENS) with lots of help from Wendelin Werner University of Bridge Decomposition of Restriction Measures Tom Alberts joint work with Hugo Duminil (ENS) with lots of help from Wendelin Werner University of Toronto Self-Avoiding Walk Bridges Self-Avoiding Walk Bridges

More information

One-factor ANOVA by example

One-factor ANOVA by example ANOVA One-factor ANOVA by example 2 One-factor ANOVA by visual inspection 3 4 One-factor ANOVA H 0 H 0 : µ 1 = µ 2 = µ 3 = H A : not all means are equal 5 One-factor ANOVA but why not t-tests t-tests?

More information

Random Coefficient Models for Time-Series Cross-Section Data: Monte Carlo Experiments of Finite Sample Properties

Random Coefficient Models for Time-Series Cross-Section Data: Monte Carlo Experiments of Finite Sample Properties Random Coefficient Models for Time-Series Cross-Section Data: Monte Carlo Experiments of Finite Sample Properties Nathaniel Beck Jonathan N. Katz Draft of April 1, 2006 We gratefully acknowledge the financial

More information

Safety Assessment of Installing Traffic Signals at High-Speed Expressway Intersections

Safety Assessment of Installing Traffic Signals at High-Speed Expressway Intersections Safety Assessment of Installing Traffic Signals at High-Speed Expressway Intersections Todd Knox Center for Transportation Research and Education Iowa State University 2901 South Loop Drive, Suite 3100

More information

Calibration and Transferability of Accident Prediction Models for Urban Intersections

Calibration and Transferability of Accident Prediction Models for Urban Intersections Calibration and Transferability of Accident Prediction Models for Urban Intersections Bhagwant Persaud Department of Civil Engineering Ryerson University 350 Victoria Street, Toronto M5B2K3 Phone: 416-979-5345;

More information

Hierarchical Bayesian Modeling of Hitting Performance in Baseball

Hierarchical Bayesian Modeling of Hitting Performance in Baseball Bayesian Analysis (2004) 1, Number 1, pp. 1 17 Hierarchical Bayesian Modeling of Hitting Performance in Baseball Shane T. Jensen Department of Statistics The Wharton School University of Pennsylvania stjensen@wharton.upenn.edu

More information

Finding your feet: modelling the batting abilities of cricketers using Gaussian processes

Finding your feet: modelling the batting abilities of cricketers using Gaussian processes Finding your feet: modelling the batting abilities of cricketers using Gaussian processes Oliver Stevenson & Brendon Brewer PhD candidate, Department of Statistics, University of Auckland o.stevenson@auckland.ac.nz

More information

Analysis of Variance. Copyright 2014 Pearson Education, Inc.

Analysis of Variance. Copyright 2014 Pearson Education, Inc. Analysis of Variance 12-1 Learning Outcomes Outcome 1. Understand the basic logic of analysis of variance. Outcome 2. Perform a hypothesis test for a single-factor design using analysis of variance manually

More information

JPEG-Compatibility Steganalysis Using Block-Histogram of Recompression Artifacts

JPEG-Compatibility Steganalysis Using Block-Histogram of Recompression Artifacts JPEG-Compatibility Steganalysis Using Block-Histogram of Recompression Artifacts Jan Kodovský, Jessica Fridrich May 16, 2012 / IH Conference 1 / 19 What is JPEG-compatibility steganalysis? Detects embedding

More information

The Intrinsic Value of a Batted Ball Technical Details

The Intrinsic Value of a Batted Ball Technical Details The Intrinsic Value of a Batted Ball Technical Details Glenn Healey, EECS Department University of California, Irvine, CA 9617 Given a set of observed batted balls and their outcomes, we develop a method

More information

Waseda Statistical Symposium on Time Series and Related Topics. ---A Satellite Meeting of IMS-APRM

Waseda Statistical Symposium on Time Series and Related Topics. ---A Satellite Meeting of IMS-APRM Waseda Statistical Symposium on Time Series and Related Topics ---A Satellite Meeting of IMS-APRM 2012--- Date : 5 7 July, 2012 Venue : Waseda University Waseda campus, International conference center

More information

Safety Evaluation of Intersection Safety Camera Program in Edmonton

Safety Evaluation of Intersection Safety Camera Program in Edmonton Safety Evaluation of Intersection Safety Camera Program in Edmonton by Laura Contini A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science In Transportation

More information

Stats 2002: Probabilities for Wins and Losses of Online Gambling

Stats 2002: Probabilities for Wins and Losses of Online Gambling Abstract: Jennifer Mateja Andrea Scisinger Lindsay Lacher Stats 2002: Probabilities for Wins and Losses of Online Gambling The objective of this experiment is to determine whether online gambling is a

More information

International Journal of Research in Science and Technology Volume 1, Issue 1: October - December, 2014

International Journal of Research in Science and Technology Volume 1, Issue 1: October - December, 2014 AN APPLICATION OF MULTILAYER PERCEPTRON NEURAL NETWORK TO PREDICT THE PERFORMANCE OF BATSMEN IN INDIAN PREMIER LEAGUE Hemanta Saikia 1 and Dibyojyoti Bhattacharjee 2 Assistant 1 Professor, The Assam Kaziranga

More information

Updated and revised standardized catch rate of blue sharks caught by the Taiwanese longline fishery in the Indian Ocean

Updated and revised standardized catch rate of blue sharks caught by the Taiwanese longline fishery in the Indian Ocean Updated and revised standardized catch rate of blue sharks caught by the Taiwanese longline fishery in the Indian Ocean Wen-Pei Tsai 1,3 and Kwang-Ming Liu 2 1 Department of Fisheries Production and Management,

More information

Calculation of Trail Usage from Counter Data

Calculation of Trail Usage from Counter Data 1. Introduction 1 Calculation of Trail Usage from Counter Data 1/17/17 Stephen Martin, Ph.D. Automatic counters are used on trails to measure how many people are using the trail. A fundamental question

More information

Modelling the distribution of first innings runs in T20 Cricket

Modelling the distribution of first innings runs in T20 Cricket Modelling the distribution of first innings runs in T20 Cricket James Kirkby The joy of smoothing James Kirkby Modelling the distribution of first innings runs in T20 Cricket The joy of smoothing 1 / 22

More information

Introduction to Machine Learning NPFL 054

Introduction to Machine Learning NPFL 054 Introduction to Machine Learning NPFL 054 http://ufal.mff.cuni.cz/course/npfl054 Barbora Hladká hladka@ufal.mff.cuni.cz Martin Holub holub@ufal.mff.cuni.cz Charles University, Faculty of Mathematics and

More information

Software Reliability 1

Software Reliability 1 Software Reliability 1 Software Reliability What is software reliability? the probability of failure-free software operation for a specified period of time in a specified environment input sw output We

More information

A Characterization of Equilibrium Set of Persuasion Games with Binary Actions

A Characterization of Equilibrium Set of Persuasion Games with Binary Actions A Characterization of Equilibrium Set of Persuasion Games with Binary Actions Shintaro Miura February 7, 2014 Abstract This paper considers a persuasion game between one sender and one receiver. The perfectly

More information

Standardized catch rates of Atlantic king mackerel (Scomberomorus cavalla) from the North Carolina Commercial fisheries trip ticket.

Standardized catch rates of Atlantic king mackerel (Scomberomorus cavalla) from the North Carolina Commercial fisheries trip ticket. SEDAR16 DW 11 Standardized catch rates of Atlantic king mackerel (Scomberomorus cavalla) from the North Carolina Commercial fisheries trip ticket. Alan Bianchi 1 and Mauricio Ortiz 2 SUMMARY Standardized

More information

Creel Survey Sampling Designs for Estimating Effort in Short-Duration Chinook Salmon Fisheries

Creel Survey Sampling Designs for Estimating Effort in Short-Duration Chinook Salmon Fisheries This article was downloaded by: [Joshua L. McCormick] On: 25 September 213, At: 7:7 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 172954 Registered office:

More information

a) List and define all assumptions for multiple OLS regression. These are all listed in section 6.5

a) List and define all assumptions for multiple OLS regression. These are all listed in section 6.5 Prof. C. M. Dalton ECN 209A Spring 2015 Practice Problems (After HW1, HW2, before HW3) CORRECTED VERSION Question 1. Draw and describe a relationship with heteroskedastic errors. Support your claim with

More information

Fundamentals of Machine Learning for Predictive Data Analytics

Fundamentals of Machine Learning for Predictive Data Analytics Fundamentals of Machine Learning for Predictive Data Analytics Appendix A Descriptive Statistics and Data Visualization for Machine learning John Kelleher and Brian Mac Namee and Aoife D Arcy john.d.kelleher@dit.ie

More information

A point-based Bayesian hierarchical model to predict the outcome of tennis matches

A point-based Bayesian hierarchical model to predict the outcome of tennis matches A point-based Bayesian hierarchical model to predict the outcome of tennis matches Martin Ingram, Silverpond September 21, 2017 Introduction Predicting tennis matches is of interest for a number of applications:

More information

Running head: DATA ANALYSIS AND INTERPRETATION 1

Running head: DATA ANALYSIS AND INTERPRETATION 1 Running head: DATA ANALYSIS AND INTERPRETATION 1 Data Analysis and Interpretation Final Project Vernon Tilly Jr. University of Central Oklahoma DATA ANALYSIS AND INTERPRETATION 2 Owners of the various

More information

Tutorial for the. Total Vertical Uncertainty Analysis Tool in NaviModel3

Tutorial for the. Total Vertical Uncertainty Analysis Tool in NaviModel3 Tutorial for the Total Vertical Uncertainty Analysis Tool in NaviModel3 May, 2011 1. Introduction The Total Vertical Uncertainty Analysis Tool in NaviModel3 has been designed to facilitate a determination

More information

The Credible Diversity Plot: A Graphic for the Comparison of Biodiversity

The Credible Diversity Plot: A Graphic for the Comparison of Biodiversity The Credible Diversity Plot: A Graphic for the Comparison of Biodiversity Christopher J. Mecklin Department of Mathematics & Statistics Murray State University Murray KY 42071 E-mail: christopher.mecklin@murraystate.edu

More information

Bayesball: A Bayesian Hierarchical Model for Evaluating Fielding in Major League Baseball

Bayesball: A Bayesian Hierarchical Model for Evaluating Fielding in Major League Baseball University of Pennsylvania ScholarlyCommons Statistics Papers Wharton Faculty Research 2009 Bayesball: A Bayesian Hierarchical Model for Evaluating Fielding in Major League Baseball Shane T. Jensen University

More information

The collapse of cod in Eastern Canada: the evidence from tagging data

The collapse of cod in Eastern Canada: the evidence from tagging data ICES Journal of Marine Science, 53: 629 640. 1996 The collapse of cod in Eastern Canada: the evidence from tagging data R. A. Myers, N. J. Barrowman, J. M. Hoenig, and Z. Qu Myers, R. A., Barrowman, N.

More information

Stock Return Quantiles and Conditional Asymmetry. - An Approach to Portfolio Selection

Stock Return Quantiles and Conditional Asymmetry. - An Approach to Portfolio Selection Stock Return Quantiles and Conditional Asymmetry - An Approach to Portfolio Selection Hanwei Liu March 6, 2017 Abstract Fully aware of the importance of effective risk management, we develop a dynamic

More information

An Empirical Comparison of Regression Analysis Strategies with Discrete Ordinal Variables

An Empirical Comparison of Regression Analysis Strategies with Discrete Ordinal Variables Kromrey & Rendina-Gobioff An Empirical Comparison of Regression Analysis Strategies with Discrete Ordinal Variables Jeffrey D. Kromrey Gianna Rendina-Gobioff University of South Florida The Type I error

More information

Do New Bike Share Stations Increase Member Use?: A Quasi-Experimental Study

Do New Bike Share Stations Increase Member Use?: A Quasi-Experimental Study Do New Bike Share Stations Increase Member Use?: A Quasi-Experimental Study Jueyu Wang & Greg Lindsey Humphrey School of Public Affairs University of Minnesota Acknowledgement: NSF Sustainable Research

More information

Special Topics: Data Science

Special Topics: Data Science Special Topics: Data Science L Linear Methods for Prediction Dr. Vidhyasaharan Sethu School of Electrical Engineering & Telecommunications University of New South Wales Sydney, Australia V. Sethu 1 Topics

More information

Wave Modeling and Applications

Wave Modeling and Applications Wave Modeling and Applications Hellenic Naval Academy Section of Mathematics Joint work with - Hellenic Navy Hydrographic Service - Atmospheric Modeling Group, Department of Physics, University of Athens,

More information

D1.2 REPORT ON MOTORCYCLISTS IMPACTS WITH ROAD INFRASTRUCTURE BASED OF AN INDEPTH INVESTIGATION OF MOTORCYCLE ACCIDENTS

D1.2 REPORT ON MOTORCYCLISTS IMPACTS WITH ROAD INFRASTRUCTURE BASED OF AN INDEPTH INVESTIGATION OF MOTORCYCLE ACCIDENTS WP 1 D1.2 REPORT ON MOTORCYCLISTS IMPACTS WITH ROAD INFRASTRUCTURE BASED OF AN INDEPTH INVESTIGATION OF MOTORCYCLE ACCIDENTS Project Acronym: Smart RRS Project Full Title: Innovative Concepts for smart

More information