SlideShare a Scribd company logo
Toward a unified approach to fitting loss models Jacques Rioux and Stuart Klugman, for presentation at the IAC, Feb. 9, 2004
Overview ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
The problem ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
The general idea ,[object Object],[object Object],[object Object]
Distributions ,[object Object],[object Object],[object Object],[object Object]
A few familiar distributions ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Flexible ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Why mixtures? ,[object Object],[object Object]
Estimating parameters ,[object Object],[object Object],[object Object],[object Object]
Representing the data ,[object Object],[object Object],[object Object]
What is the issue? ,[object Object],[object Object],[object Object],[object Object],[object Object]
Issue – grouped data ,[object Object],[object Object],[object Object]
Review ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Example ,[object Object],[object Object],[object Object]
Empirical cdf
Distribution function plot ,[object Object],[object Object]
Example model ,[object Object]
Distribution function plot
Confidence bands ,[object Object],[object Object]
CDF plot with bounds
Other CDF pictures ,[object Object],[object Object]
CDF difference plot
Histogram plot ,[object Object],[object Object]
Histogram plot
Hypothesis tests ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Kolmogorov-Smirnov ,[object Object],[object Object]
Anderson-Darling ,[object Object],[object Object],[object Object],[object Object]
Chi-square test ,[object Object],[object Object]
Results ,[object Object],[object Object],[object Object],[object Object]
Comparing models ,[object Object],[object Object],[object Object]
Several models Model Loglike A-D K-S Chi-sq SBC Exp -628.23 1.2245 0.9739 0.1054 -630.53 Ln -626.26 0.6682 0.9375 0.2126 -630.87 Gam -627.35 0.8369 1.0355 0.2319 -631.96 L/E -623.77 0.2579 0.5829 0.5608 -632.98 G/E -623.64 0.2804 0.5773 0.5260 -632.85 L/E/E -623.39 0.1484 0.4494 0.3472 -637.21 G/E/E -623.26 0.1353 0.4652 0.3348 -637.08
Which is the winner? ,[object Object],[object Object],[object Object],[object Object],[object Object]
Can this be automated? ,[object Object],[object Object],[object Object],[object Object],[object Object]

More Related Content

What's hot

slides
slidesslides
slides
butest
 
Chapter8
Chapter8Chapter8
Dataa miining
Dataa miiningDataa miining
Dataa miining
SUBBIAH SURESH
 
Instance based learning
Instance based learningInstance based learning
Instance based learning
swapnac12
 
lecture_mooney.ppt
lecture_mooney.pptlecture_mooney.ppt
lecture_mooney.ppt
butest
 
Types of clustering and different types of clustering algorithms
Types of clustering and different types of clustering algorithmsTypes of clustering and different types of clustering algorithms
Types of clustering and different types of clustering algorithms
Prashanth Guntal
 
Clustering
ClusteringClustering
Clustering
Meme Hei
 
lab report 4
lab report 4lab report 4
lab report 4
Selase Kwami
 
Lec4 Clustering
Lec4 ClusteringLec4 Clustering
Lec4 Clustering
Jeff Hammerbacher
 
Unsupervised learning clustering
Unsupervised learning clusteringUnsupervised learning clustering
Unsupervised learning clustering
Dr Nisha Arora
 
Clustering (from Google)
Clustering (from Google)Clustering (from Google)
Clustering (from Google)
Sri Prasanna
 
Selection K in K-means Clustering
Selection K in K-means ClusteringSelection K in K-means Clustering
Selection K in K-means Clustering
Junghoon Kim
 
Af4201214217
Af4201214217Af4201214217
Af4201214217
IJERA Editor
 
Clustering
ClusteringClustering
Parameter Optimisation for Automated Feature Point Detection
Parameter Optimisation for Automated Feature Point DetectionParameter Optimisation for Automated Feature Point Detection
Parameter Optimisation for Automated Feature Point Detection
Dario Panada
 
Clustering
ClusteringClustering
Clustering
LipikaSaha2
 
Pillar k means
Pillar k meansPillar k means
Pillar k means
swathi b
 
Nearest neighbors
Nearest neighborsNearest neighbors
Nearest neighbors
zekeLabs Technologies
 
Big data Clustering Algorithms And Strategies
Big data Clustering Algorithms And StrategiesBig data Clustering Algorithms And Strategies
Big data Clustering Algorithms And Strategies
Farzad Nozarian
 
Cure, Clustering Algorithm
Cure, Clustering AlgorithmCure, Clustering Algorithm
Cure, Clustering Algorithm
Lino Possamai
 

What's hot (20)

slides
slidesslides
slides
 
Chapter8
Chapter8Chapter8
Chapter8
 
Dataa miining
Dataa miiningDataa miining
Dataa miining
 
Instance based learning
Instance based learningInstance based learning
Instance based learning
 
lecture_mooney.ppt
lecture_mooney.pptlecture_mooney.ppt
lecture_mooney.ppt
 
Types of clustering and different types of clustering algorithms
Types of clustering and different types of clustering algorithmsTypes of clustering and different types of clustering algorithms
Types of clustering and different types of clustering algorithms
 
Clustering
ClusteringClustering
Clustering
 
lab report 4
lab report 4lab report 4
lab report 4
 
Lec4 Clustering
Lec4 ClusteringLec4 Clustering
Lec4 Clustering
 
Unsupervised learning clustering
Unsupervised learning clusteringUnsupervised learning clustering
Unsupervised learning clustering
 
Clustering (from Google)
Clustering (from Google)Clustering (from Google)
Clustering (from Google)
 
Selection K in K-means Clustering
Selection K in K-means ClusteringSelection K in K-means Clustering
Selection K in K-means Clustering
 
Af4201214217
Af4201214217Af4201214217
Af4201214217
 
Clustering
ClusteringClustering
Clustering
 
Parameter Optimisation for Automated Feature Point Detection
Parameter Optimisation for Automated Feature Point DetectionParameter Optimisation for Automated Feature Point Detection
Parameter Optimisation for Automated Feature Point Detection
 
Clustering
ClusteringClustering
Clustering
 
Pillar k means
Pillar k meansPillar k means
Pillar k means
 
Nearest neighbors
Nearest neighborsNearest neighbors
Nearest neighbors
 
Big data Clustering Algorithms And Strategies
Big data Clustering Algorithms And StrategiesBig data Clustering Algorithms And Strategies
Big data Clustering Algorithms And Strategies
 
Cure, Clustering Algorithm
Cure, Clustering AlgorithmCure, Clustering Algorithm
Cure, Clustering Algorithm
 

Viewers also liked

Моніторинг ЦНАПів у Львові: результати та зміни. Наталя Міхнова
Моніторинг ЦНАПів у Львові: результати та зміни. Наталя МіхноваМоніторинг ЦНАПів у Львові: результати та зміни. Наталя Міхнова
Моніторинг ЦНАПів у Львові: результати та зміни. Наталя Міхнова
Galyna Smirnova
 
Na Col De Brusel·Les Es Perd
Na Col De Brusel·Les Es PerdNa Col De Brusel·Les Es Perd
Na Col De Brusel·Les Es PerdPili Villalobos
 
HCR Alert jun_2012_guidance_fsa_limit
HCR Alert jun_2012_guidance_fsa_limitHCR Alert jun_2012_guidance_fsa_limit
HCR Alert jun_2012_guidance_fsa_limit
Annette Wright, GBA, GBDS
 
E107 Open Education Practice and Potential: Session 1
E107 Open Education Practice and Potential: Session 1E107 Open Education Practice and Potential: Session 1
E107 Open Education Practice and Potential: Session 1
Brandon Muramatsu
 
Beba
BebaBeba
Intro to Using Commons Groups for Internal Communications
Intro to Using Commons Groups for Internal CommunicationsIntro to Using Commons Groups for Internal Communications
Intro to Using Commons Groups for Internal Communications
Diana Grappasonno
 

Viewers also liked (6)

Моніторинг ЦНАПів у Львові: результати та зміни. Наталя Міхнова
Моніторинг ЦНАПів у Львові: результати та зміни. Наталя МіхноваМоніторинг ЦНАПів у Львові: результати та зміни. Наталя Міхнова
Моніторинг ЦНАПів у Львові: результати та зміни. Наталя Міхнова
 
Na Col De Brusel·Les Es Perd
Na Col De Brusel·Les Es PerdNa Col De Brusel·Les Es Perd
Na Col De Brusel·Les Es Perd
 
HCR Alert jun_2012_guidance_fsa_limit
HCR Alert jun_2012_guidance_fsa_limitHCR Alert jun_2012_guidance_fsa_limit
HCR Alert jun_2012_guidance_fsa_limit
 
E107 Open Education Practice and Potential: Session 1
E107 Open Education Practice and Potential: Session 1E107 Open Education Practice and Potential: Session 1
E107 Open Education Practice and Potential: Session 1
 
Beba
BebaBeba
Beba
 
Intro to Using Commons Groups for Internal Communications
Intro to Using Commons Groups for Internal CommunicationsIntro to Using Commons Groups for Internal Communications
Intro to Using Commons Groups for Internal Communications
 

Similar to Toward a Unified Approach to Fitting Loss Models

Intro to Model Selection
Intro to Model SelectionIntro to Model Selection
Intro to Model Selection
chenhm
 
German credit score shivaram prakash
German credit score shivaram prakashGerman credit score shivaram prakash
German credit score shivaram prakash
Shivaram Prakash
 
Model selection
Model selectionModel selection
Model selection
Animesh Kumar
 
Statistical Data Analysis on a Data Set (Diabetes 130-US hospitals for years ...
Statistical Data Analysis on a Data Set (Diabetes 130-US hospitals for years ...Statistical Data Analysis on a Data Set (Diabetes 130-US hospitals for years ...
Statistical Data Analysis on a Data Set (Diabetes 130-US hospitals for years ...
Seval Çapraz
 
Max Kuhn's talk on R machine learning
Max Kuhn's talk on R machine learningMax Kuhn's talk on R machine learning
Max Kuhn's talk on R machine learning
Vivian S. Zhang
 
Guide for building GLMS
Guide for building GLMSGuide for building GLMS
Guide for building GLMS
Ali T. Lotia
 
Data Science - Part V - Decision Trees & Random Forests
Data Science - Part V - Decision Trees & Random Forests Data Science - Part V - Decision Trees & Random Forests
Data Science - Part V - Decision Trees & Random Forests
Derek Kane
 
November, 2006 CCKM'06 1
November, 2006 CCKM'06 1 November, 2006 CCKM'06 1
November, 2006 CCKM'06 1
butest
 
Lecture 17: Supervised Learning Recap
Lecture 17: Supervised Learning RecapLecture 17: Supervised Learning Recap
Lecture 17: Supervised Learning Recap
butest
 
Probability density estimation using Product of Conditional Experts
Probability density estimation using Product of Conditional ExpertsProbability density estimation using Product of Conditional Experts
Probability density estimation using Product of Conditional Experts
Chirag Gupta
 
Multiple Regression
Multiple RegressionMultiple Regression
Multiple Regression
Nicholas Manurung
 
Diabetes data - model assessment using R
Diabetes data - model assessment using RDiabetes data - model assessment using R
Diabetes data - model assessment using R
Gregg Barrett
 
P & C Reserving Using GAMLSS
P & C Reserving Using GAMLSSP & C Reserving Using GAMLSS
P & C Reserving Using GAMLSS
Giorgio Alfredo Spedicato
 
report
reportreport
report
Arthur He
 
Model Selection Techniques
Model Selection TechniquesModel Selection Techniques
Model Selection Techniques
Swati .
 
SVM - Functional Verification
SVM - Functional VerificationSVM - Functional Verification
SVM - Functional Verification
Sai Kiran Kadam
 
Auto MPG Regression Analysis
Auto MPG Regression AnalysisAuto MPG Regression Analysis
Auto MPG Regression Analysis
Anirudh Srinath.V
 
[CIKM 2014] Deviation-Based Contextual SLIM Recommenders
[CIKM 2014] Deviation-Based Contextual SLIM Recommenders[CIKM 2014] Deviation-Based Contextual SLIM Recommenders
[CIKM 2014] Deviation-Based Contextual SLIM Recommenders
YONG ZHENG
 
AUTO MPG Regression Analysis
AUTO MPG Regression AnalysisAUTO MPG Regression Analysis
AUTO MPG Regression Analysis
ShankarPrasaadRajama
 
07 dimensionality reduction
07 dimensionality reduction07 dimensionality reduction
07 dimensionality reduction
Marco Quartulli
 

Similar to Toward a Unified Approach to Fitting Loss Models (20)

Intro to Model Selection
Intro to Model SelectionIntro to Model Selection
Intro to Model Selection
 
German credit score shivaram prakash
German credit score shivaram prakashGerman credit score shivaram prakash
German credit score shivaram prakash
 
Model selection
Model selectionModel selection
Model selection
 
Statistical Data Analysis on a Data Set (Diabetes 130-US hospitals for years ...
Statistical Data Analysis on a Data Set (Diabetes 130-US hospitals for years ...Statistical Data Analysis on a Data Set (Diabetes 130-US hospitals for years ...
Statistical Data Analysis on a Data Set (Diabetes 130-US hospitals for years ...
 
Max Kuhn's talk on R machine learning
Max Kuhn's talk on R machine learningMax Kuhn's talk on R machine learning
Max Kuhn's talk on R machine learning
 
Guide for building GLMS
Guide for building GLMSGuide for building GLMS
Guide for building GLMS
 
Data Science - Part V - Decision Trees & Random Forests
Data Science - Part V - Decision Trees & Random Forests Data Science - Part V - Decision Trees & Random Forests
Data Science - Part V - Decision Trees & Random Forests
 
November, 2006 CCKM'06 1
November, 2006 CCKM'06 1 November, 2006 CCKM'06 1
November, 2006 CCKM'06 1
 
Lecture 17: Supervised Learning Recap
Lecture 17: Supervised Learning RecapLecture 17: Supervised Learning Recap
Lecture 17: Supervised Learning Recap
 
Probability density estimation using Product of Conditional Experts
Probability density estimation using Product of Conditional ExpertsProbability density estimation using Product of Conditional Experts
Probability density estimation using Product of Conditional Experts
 
Multiple Regression
Multiple RegressionMultiple Regression
Multiple Regression
 
Diabetes data - model assessment using R
Diabetes data - model assessment using RDiabetes data - model assessment using R
Diabetes data - model assessment using R
 
P & C Reserving Using GAMLSS
P & C Reserving Using GAMLSSP & C Reserving Using GAMLSS
P & C Reserving Using GAMLSS
 
report
reportreport
report
 
Model Selection Techniques
Model Selection TechniquesModel Selection Techniques
Model Selection Techniques
 
SVM - Functional Verification
SVM - Functional VerificationSVM - Functional Verification
SVM - Functional Verification
 
Auto MPG Regression Analysis
Auto MPG Regression AnalysisAuto MPG Regression Analysis
Auto MPG Regression Analysis
 
[CIKM 2014] Deviation-Based Contextual SLIM Recommenders
[CIKM 2014] Deviation-Based Contextual SLIM Recommenders[CIKM 2014] Deviation-Based Contextual SLIM Recommenders
[CIKM 2014] Deviation-Based Contextual SLIM Recommenders
 
AUTO MPG Regression Analysis
AUTO MPG Regression AnalysisAUTO MPG Regression Analysis
AUTO MPG Regression Analysis
 
07 dimensionality reduction
07 dimensionality reduction07 dimensionality reduction
07 dimensionality reduction
 

Toward a Unified Approach to Fitting Loss Models

  • 1. Toward a unified approach to fitting loss models Jacques Rioux and Stuart Klugman, for presentation at the IAC, Feb. 9, 2004
  • 2.
  • 3.
  • 4.
  • 5.
  • 6.
  • 7.
  • 8.
  • 9.
  • 10.
  • 11.
  • 12.
  • 13.
  • 14.
  • 16.
  • 17.
  • 19.
  • 20. CDF plot with bounds
  • 21.
  • 23.
  • 25.
  • 26.
  • 27.
  • 28.
  • 29.
  • 30.
  • 31. Several models Model Loglike A-D K-S Chi-sq SBC Exp -628.23 1.2245 0.9739 0.1054 -630.53 Ln -626.26 0.6682 0.9375 0.2126 -630.87 Gam -627.35 0.8369 1.0355 0.2319 -631.96 L/E -623.77 0.2579 0.5829 0.5608 -632.98 G/E -623.64 0.2804 0.5773 0.5260 -632.85 L/E/E -623.39 0.1484 0.4494 0.3472 -637.21 G/E/E -623.26 0.1353 0.4652 0.3348 -637.08
  • 32.
  • 33.