Probabilities Probability Distribution Predictor Variables Prior Information New Data Prior and New Data Overview.

Slides:



Advertisements
Similar presentations
Bayes rule, priors and maximum a posteriori
Advertisements

Bivariate Normal Distribution and Regression Application to Galton’s Heights of Adult Children and Parents Sources: Galton, Francis (1889). Natural Inheritance,
Naive Bayes Classifiers, an Overview By Roozmehr Safi.
Notes Sample vs distribution “m” vs “µ” and “s” vs “σ” Bias/Variance Bias: Measures how much the learnt model is wrong disregarding noise Variance: Measures.
INTRODUCTION TO MACHINE LEARNING Bayesian Estimation.
Ch5 Stochastic Methods Dr. Bernard Chen Ph.D. University of Central Arkansas Spring 2011.
EXAMPLE 1 Construct a probability distribution
EXAMPLE 1 Construct a probability distribution Let X be a random variable that represents the sum when two six-sided dice are rolled. Make a table and.
Naïve Bayes Classifier
Assuming normally distributed data! Naïve Bayes Classifier.
Ridge regression and Bayesian linear regression Kenneth D. Harris 6/5/15.
Chapter 8 Logistic Regression 1. Introduction Logistic regression extends the ideas of linear regression to the situation where the dependent variable,
Multinomial Logistic Regression
Probability theory 2008 Conditional probability mass function  Discrete case  Continuous case.
Probability theory 2010 Conditional distributions  Conditional probability:  Conditional probability mass function: Discrete case  Conditional probability.
Simple Bayesian Supervised Models Saskia Klein & Steffen Bollmann 1.
 The Law of Large Numbers – Read the preface to Chapter 7 on page 388 and be prepared to summarize the Law of Large Numbers.
1 Linear Classification Problem Two approaches: -Fisher’s Linear Discriminant Analysis -Logistic regression model.
Logistic regression for binary response variables.
Correlation and Covariance. Overview Continuous Categorical Histogram Scatter Boxplot Predictor Variable (X-Axis) Height Outcome, Dependent Variable (Y-Axis)
1 G Lect 11W Logistic Regression Review Maximum Likelihood Estimates Probit Regression and Example Model Fit G Multiple Regression Week 11.
PBG 650 Advanced Plant Breeding
Simple Mathematical Facts for Lecture 1. Conditional Probabilities Given an event has occurred, the conditional probability that another event occurs.
1 1 Slide © 2008 Thomson South-Western. All Rights Reserved Chapter 15 Multiple Regression n Multiple Regression Model n Least Squares Method n Multiple.
Learning Theory Reza Shadmehr logistic regression, iterative re-weighted least squares.
Section 5.2: Linear Regression: Fitting a Line to Bivariate Data.
Business Intelligence and Decision Modeling Week 11 Predictive Modeling (2) Logistic Regression.
A Brief History of Statistics. Medieval Times: Dice and Gambling.
Empirical Research Methods in Computer Science Lecture 7 November 30, 2005 Noah Smith.
Learning Theory Reza Shadmehr Linear and quadratic decision boundaries Kernel estimates of density Missing data.
Probability Serena Saliba. What is probability? Probability is a measure of how likely it is that some event will occur. Examples: The odds of winning.
Introduction to Regression Analysis. Dependent variable (response variable) Measures an outcome of a study  Income  GRE scores Dependent variable =
What is Probability?. The Mathematics of Chance How many possible outcomes are there with a single 6-sided die? What are your “chances” of rolling a 6?
College Prep Stats. x is the independent variable (predictor variable) ^ y = b 0 + b 1 x ^ y = mx + b b 0 = y - intercept b 1 = slope y is the dependent.
SW318 Social Work Statistics Slide 1 Logistic Regression and Odds Ratios Example of Odds Ratio Using Relationship between Death Penalty and Race.
Discrete Distributions. Random Variable - A numerical variable whose value depends on the outcome of a chance experiment.
XIAO WU DATA ANALYSIS & BASIC STATISTICS.
Where to Get Data? Run an Experiment Use Existing Data.
What factors are most responsible for height?. Model Specification ERROR??? measurement error model error analysis unexplained unknown unaccounted for.
Logistic regression (when you have a binary response variable)
Steps Continuous Categorical Histogram Scatter Boxplot Child’s Height Linear Regression Dad’s Height Gender Continuous Y X1, X2 X3 Type Variable Mom’s.
ROLL A PAIR OF DICE AND ADD THE NUMBERS Possible Outcomes: There are 6 x 6 = 36 equally likely.
Continuous Outcome, Dependent Variable (Y-Axis) Child’s Height
Construction Engineering 221 Probability and Statistics.
Probability Distributions Section 7.6. Definitions Random Variable: values are numbers determined by the outcome of an experiment. (rolling 2 dice: rv’s.
Cavalier De Merè was a French nobleman who had a hobby: gambling. A day he wondered if it was more likely to get a 6, throwing a dice 4 times or to get.
Conditional Probability 423/what-is-your-favorite-data-analysis-cartoon 1.
Logistic Regression When and why do we use logistic regression?
Simple Linear Regression
Chapter 5 Review MDM 4U Mr. Lieff.
Chapter 5 Review MDM 4U Gary Greer.
The Skinny on High School
CH 5: Multivariate Methods
Regression.
Data Mining Lecture 11.
Simple Linear Regression
Correlation and regression Log. Reg
OVERVIEW OF BAYESIAN INFERENCE: PART 1
Suppose you roll two dice, and let X be sum of the dice. Then X is
Probability Review for Financial Engineers
Is a persons’ size related to if they were bullied
The Binomial Distribution
Mathematical Foundations of BME Reza Shadmehr
Probability Key Questions
Binomial Distributions
E370 Statistical analysis for bus & econ
Probability The risk of getting struck by lightning in any year is 1 in 750,000. The chances of surviving a lightning strike are 3 in 4. These risks.
Mathematical Foundations of BME
Mathematical Foundations of BME
Mathematical Foundations of BME Reza Shadmehr
Presentation transcript:

Probabilities Probability Distribution Predictor Variables Prior Information New Data Prior and New Data Overview

Medieval Times: Dice and Gambling

Modern Times: Dice and Games/Gambing

Dice Probabilities 1616 =16.7% = 2.78% 6 36 =16.78% Dice Outcome are Independent Sum

Dice Probabilities Probability Distribution

Blaise Pascal 1600’s: Probability & Gambling one "6" in four rolls one double-six in 24 throws Do these have equal probabilities? Chevalier de Méré

Prediction Model: Dice 1616 =16.7% Y = ? No Predictor Variables

Prediction Model: Heights ChildHeight = FatherHeight + MotherHeight + Gender + Ɛ Predictor Variables!!! Linear Regression invented in 1877 by Francis Galton

Prediction Model: Logistic Logistic Regression invented in 1838 by Pierre-Francois Verhulst

Probability & Classification: Gender ~ Height Let’s Invert the Problem – “Given Child Height What is the Gender?” and Pretend its 1761 – Before Logistic Regression Gender ChildHeight (Categorical)(Continuous)

1761: Bayesian Probability Distribution New Data Probability Female Probability Male Height of the Person = Data Prior (X) Data Prior (X) = Gender Prior (X) Child Height 66.5

Bayesian Formulas Same for both female and male

Normal Distribution and Probability D D

Bayesian Formulas D D D

Bayesian Formulas – Excel D

Naïve Bayes 84.1%

Naïve Bayes

Probability: Gender ~ Height + Weight + FootSize