By Sharath Kumar Aitha. Instructor: Dr. Dongchul Kim.

Slides:



Advertisements
Similar presentations
Pattern Recognition and Machine Learning
Advertisements

Dummy Dependent variable Models
Prof. Navneet Goyal CS & IS BITS, Pilani
GENERAL LINEAR MODELS: Estimation algorithms
Unsupervised Learning
Logistic Regression.
6-1 Introduction To Empirical Models 6-1 Introduction To Empirical Models.
The General Linear Model. The Simple Linear Model Linear Regression.
Nguyen Ngoc Anh Nguyen Ha Trang
Chapter 4: Linear Models for Classification
Logistic Regression STA302 F 2014 See last slide for copyright information 1.
Maximum likelihood Conditional distribution and likelihood Maximum likelihood estimations Information in the data and likelihood Observed and Fisher’s.
Introduction to Logistic Regression. Simple linear regression Table 1 Age and systolic blood pressure (SBP) among 33 adult women.
Descriptive statistics Experiment  Data  Sample Statistics Experiment  Data  Sample Statistics Sample mean Sample mean Sample variance Sample variance.
Kernel Methods Part 2 Bing Han June 26, Local Likelihood Logistic Regression.
Multivariate Probability Distributions. Multivariate Random Variables In many settings, we are interested in 2 or more characteristics observed in experiments.
CSCI 347 / CS 4206: Data Mining Module 04: Algorithms Topic 06: Regression.
1 1 Slide Simple Linear Regression Chapter 14 BA 303 – Spring 2011.
1 1 Slide © 2008 Thomson South-Western. All Rights Reserved Slides by JOHN LOUCKS & Updated by SPIROS VELIANITIS.
Collaborative Filtering Matrix Factorization Approach
Incomplete Graphical Models Nan Hu. Outline Motivation K-means clustering Coordinate Descending algorithm Density estimation EM on unconditional mixture.
Binary Variables (1) Coin flipping: heads=1, tails=0 Bernoulli Distribution.
Efficient Model Selection for Support Vector Machines
1 Logistic Regression Adapted from: Tom Mitchell’s Machine Learning Book Evan Wei Xiang and Qiang Yang.
Random Sampling, Point Estimation and Maximum Likelihood.
1 1 Slide © 2003 Thomson/South-Western Chapter 13 Multiple Regression n Multiple Regression Model n Least Squares Method n Multiple Coefficient of Determination.
Learning Theory Reza Shadmehr logistic regression, iterative re-weighted least squares.
CIS 2033 based on Dekking et al. A Modern Introduction to Probability and Statistics Instructor Longin Jan Latecki C22: The Method of Least Squares.
Logistic Regression STA2101/442 F 2014 See last slide for copyright information.
MTH 161: Introduction To Statistics
Logistic Regression Database Marketing Instructor: N. Kumar.
CS 782 – Machine Learning Lecture 4 Linear Models for Classification  Probabilistic generative models  Probabilistic discriminative models.
Managerial Economics Demand Estimation & Forecasting.
1 1 Slide © 2014 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole.
An Asymptotic Analysis of Generative, Discriminative, and Pseudolikelihood Estimators by Percy Liang and Michael Jordan (ICML 2008 ) Presented by Lihan.
Simple Linear Regression. The term linear regression implies that  Y|x is linearly related to x by the population regression equation  Y|x =  +  x.
Chapter 13 Multiple Regression
Chapter 11 Statistical Techniques. Data Warehouse and Data Mining Chapter 11 2 Chapter Objectives  Understand when linear regression is an appropriate.
Chapter 7 Point Estimation of Parameters. Learning Objectives Explain the general concepts of estimating Explain important properties of point estimators.
CSE 5331/7331 F'07© Prentice Hall1 CSE 5331/7331 Fall 2007 Machine Learning Margaret H. Dunham Department of Computer Science and Engineering Southern.
Multiple Logistic Regression STAT E-150 Statistical Methods.
1 1 Slide © 2011 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole.
Week 41 How to find estimators? There are two main methods for finding estimators: 1) Method of moments. 2) The method of Maximum likelihood. Sometimes.
Logistic Regression Analysis Gerrit Rooks
Machine Learning 5. Parametric Methods.
Logistic Regression Saed Sayad 1www.ismartsoft.com.
Additive Models , Trees , and Related Models Prof. Liqing Zhang Dept. Computer Science & Engineering, Shanghai Jiaotong University.
Statistics Sampling Distributions and Point Estimation of Parameters Contents, figures, and exercises come from the textbook: Applied Statistics and Probability.
MathematicalMarketing Slide 5.1 OLS Chapter 5: Ordinary Least Square Regression We will be discussing  The Linear Regression Model  Estimation of the.
Lecturer: Ing. Martina Hanová, PhD.. Regression analysis Regression analysis is a tool for analyzing relationships between financial variables:  Identify.
Conditional Expectation
Week 21 Statistical Model A statistical model for some data is a set of distributions, one of which corresponds to the true unknown distribution that produced.
BINARY LOGISTIC REGRESSION
Chapter 7. Classification and Prediction
By Arijit Chatterjee Dr
Probability Theory and Parameter Estimation I
Empirical risk minimization
Ch3: Model Building through Regression
Classification of unlabeled data:
John Loucks St. Edward’s University . SLIDES . BY.
Machine Learning Basics
CHAPTER 26: Inference for Regression
Ying shen Sse, tongji university Sep. 2016
POINT ESTIMATOR OF PARAMETERS
Mathematical Foundations of BME Reza Shadmehr
Simple Linear Regression
Empirical risk minimization
Multivariate Methods Berlin Chen
Multivariate Methods Berlin Chen, 2005 References:
BEC 30325: MANAGERIAL ECONOMICS
Presentation transcript:

By Sharath Kumar Aitha. Instructor: Dr. Dongchul Kim.

 Introduction.  Logistic Regression.  Formula  Conditional Density Function.  Maximum log like hood  Pseudo code  Classification  Output  Application  Conclusion

 To analyze the historical data of online shop customer behaviors by logistic predictive modeling and predicts whether the customers purchase a tablet pc online shop.  Analyzes data by datamining and establishes predictive modeling based on logistic regression.

 This is the most commonly used technology used for developing predictive modeling.  Its function is to look for a variety of equations which reflects the customer behavior patterns according to the observations of X and Y.  There are two types of regressions they are: 1.Ordinary least square Regression(OLS) 2.Logistic Regression.

 If the variable Y is continuous we usually use OLS  For example if variable Y is the sales volume of tablet PCs and the distribution of these observations is more beautiful, we can use OLS.  But the requirements of OLS of Y are very harsh and difficulty, so we use logistic regression.

 Logistic regression is a statistical method for analyzing a dataset in which there are one or more independent variables that determine an outcome.  This is a unsupervised function where we need to provide the input values.

 If we have feature data t i ∈ℜ and label data y i ∈ {0,1},i = 1,2,···,n. corresponding to two classes. we can use logistic regression to solve this by doing regression between y and t on logistic function  H(t)=1/1+e -(at-b)  T is the input.  E is the exponential  A is the coefficient to adjust the slope.  B is the parameter.

 Def: Suppose X and Y are continuous random variables with joint probability density function f(x,y) and marginal probability density functions f X (x) and f Y (y), respectively. Then, the conditional probability density function of Y given X = x is defined as: provided f X (x) > 0.  Formula: conditional density function p(y/t):  P(y/t)=h(t) y (1-h(t)) (1-y)  Which can also be written as  P(y/t)=(1/1+e -(at-b) ) y (1/1+e (at-b) ) (1-y)

 This is a statistical method used for estimating the co- efficients of a model.  The likelihood function measures the probability of observing the particular set of dependent variable values that occurs in the sample.

Algorithm 1 One dimension logistic regression Input: t; y Output: a; b 1: Initialize a and b 2: repeat 3: a = a + τ ∂ℓ/∂a 4: b = b + τ ∂ℓ/∂b 5: until convergence of a and b

 ∂ℓ/∂a  ∂ℓ/∂b

 For a testing data given feature t or x, by Eq.(1), we get h, it is the probability of y=1. so the discriminant function is:

Design and implementation:

 Logistic Regression modeling can be used in predicting the possibility of a customer to buy a commodity.  This is also helpful for the online shop owner to make the decision.