Generalised Linear Mixed Models
|
|
- Christian Marvin Allen
- 5 years ago
- Views:
Transcription
1 University of Edinburgh November 4, 2014
2 ANCOVA Bradley-Terry models MANCOVA Meta-analysis Multi-membership models Pedigree analysis: animal models Phylogenetic analysis: comparative approach Random Regression Rasch Models Regression Ridge Regression Splines Survival-analysis Threshold models Time-series Varying coefficient models
3 Outline What is a linear model? What is a random effect? MCMCglmm Non-Gaussian data Structured random effects
4 Linear Model > data(traffic, package="mass") A Swedish Experiment: On some days make everyone drive to the speed limit on others let everyone drive as fast as they want. Count how many citizens are killed.
5 Linear Model > data(traffic, package="mass") A Swedish Experiment: On some days make everyone drive to the speed limit on others let everyone drive as fast as they want. Count how many citizens are killed. > Traffic[c(1,2,184),] year day limit y no no yes 9
6 Linear Model Model Syntax y ~ limit + year + day
7 Linear Model Model Syntax y ~ limit + year + day Set of Simultaneous Equations E[y[1]] = 1β 1 + (limit[1]=="yes")β 2 + (year[1]=="1962")β 3 + day[1]β 4 E[y[2]] = 1β 1 + (limit[2]=="yes")β 2 + (year[2]=="1962")β 3 + day[2]β 4. =.. E[y[184]] = 1β 1 + (limit[184]=="yes")β 2 + (year[184]=="1962")β 3 + day[184]β 4
8 Linear Model Model Syntax y ~ limit + year + day Set of Simultaneous Equations E[y[1]] = 1β 1 + (limit[1]=="yes")β 2 + (year[1]=="1962")β 3 + day[1]β 4 E[y[2]] = 1β 1 + (limit[2]=="yes")β 2 + (year[2]=="1962")β 3 + day[2]β 4. =.. E[y[184]] = 1β 1 + (limit[184]=="yes")β 2 + (year[184]=="1962")β 3 + day[184]β 4 Compact representation: design matrix and parameter vector E[y] = Xβ
9 Linear Model Model Syntax y ~ limit + year + day Set of Simultaneous Equations E[y[1]] = 1β 1 + (limit[1]=="yes")β 2 + (year[1]=="1962")β 3 + day[1]β 4 E[y[2]] = 1β 1 + (limit[2]=="yes")β 2 + (year[2]=="1962")β 3 + day[2]β 4. =.. E[y[184]] = 1β 1 + (limit[184]=="yes")β 2 + (year[184]=="1962")β 3 + day[184]β 4 Compact representation: design matrix and parameter vector E[y] = Xβ > X<-model.matrix(y~limit+year+day, data=traffic) > X[c(1,2,184),] (Intercept) limityes year1962 day
10 Linear Model E[y] = Xβ
11 Linear Model E[y] = Xβ The full model y N(Xβ, σ 2 e I)
12 Linear Model E[y] = Xβ The full model y N(Xβ, σ 2 e I) Error structure σe 2 I = σe = σe σe σe 2
13 Linear Model > m1<-mcmcglmm(y ~ limit + year + day, data=traffic)
14 Linear Model > m1<-mcmcglmm(y ~ limit + year + day, data=traffic) > summary(m1) Iterations = 3001:12991 Thinning interval = 10 Sample size = 1000 DIC: R-structure: ~units post.mean l-95% CI u-95% CI eff.samp units Location effects: y ~ limit + year + day post.mean l-95% CI u-95% CI eff.samp pmcmc (Intercept) <0.001 *** limityes ** year day * --- Signif. codes: 0 '***' '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
15 Linear Mixed Model Random Effects: E[y[1]] = X[1, ]β E[y[2]] = X[2, ]β E[y[184] = X[184, ]β
16 Linear Mixed Model Random Effects: E[y[1]] = X[1, ]β + (day[1]=="1")u 1 + (day[1]=="2")u 2... (day[1]=="92")u 92 E[y[2]] = X[2, ]β + (day[2]=="1")u 1 + (day[2]=="2")u 2... (day[2]=="92")u 92 E[y[184] = X[184, ]β + (day[184]=="1")u 1 + (day[184]=="2")u 2... (day[184]=="92")u 92
17 Linear Mixed Model Random Effects: E[y[1]] = X[1, ]β + (day[1]=="1")u 1 + (day[1]=="2")u 2... (day[1]=="92")u 92 E[y[2]] = X[2, ]β + (day[2]=="1")u 1 + (day[2]=="2")u 2... (day[2]=="92")u 92 E[y[184] = X[184, ]β + (day[184]=="1")u 1 + (day[184]=="2")u 2... (day[184]=="92")u 92 Compact representation: design matrix and parameter vector E[y] = Xβ + Zu = Wθ
18 Linear Mixed Model Random Effects: E[y[1]] = X[1, ]β + (day[1]=="1")u 1 + (day[1]=="2")u 2... (day[1]=="92")u 92 E[y[2]] = X[2, ]β + (day[2]=="1")u 1 + (day[2]=="2")u 2... (day[2]=="92")u 92 E[y[184] = X[184, ]β + (day[184]=="1")u 1 + (day[184]=="2")u 2... (day[184]=="92")u 92 Compact representation: design matrix and parameter vector E[y] = Xβ + Zu = Wθ [ β θ = u ]
19 Linear Mixed Model Random Effects: E[y[1]] = X[1, ]β + (day[1]=="1")u 1 + (day[1]=="2")u 2... (day[1]=="92")u 92 E[y[2]] = X[2, ]β + (day[2]=="1")u 1 + (day[2]=="2")u 2... (day[2]=="92")u 92 E[y[184] = X[184, ]β + (day[184]=="1")u 1 + (day[184]=="2")u 2... (day[184]=="92")u 92 Compact representation: design matrix and parameter vector E[y] = Xβ + Zu = Wθ [ β θ = u ] W = [X, Z]
20 Linear Mixed Model Random Effects: E[y[1]] = X[1, ]β + (day[1]=="1")u 1 + (day[1]=="2")u 2... (day[1]=="92")u 92 E[y[2]] = X[2, ]β + (day[2]=="1")u 1 + (day[2]=="2")u 2... (day[2]=="92")u 92 E[y[184] = X[184, ]β + (day[184]=="1")u 1 + (day[184]=="2")u 2... (day[184]=="92")u 92 Compact representation: design matrix and parameter vector E[y] = Xβ + Zu = Wθ [ β θ = u ] W = [X, Z] > Z<-model.matrix(~as.factor(day)-1, data=traffic) > W<-cbind(X,Z)
21 Linear Mixed Model Fixed Effects σ 2 β β N(0, σ 2 βi) is not estimated, and is usually assumed to be large (or often in non-bayesian models)
22 Linear Mixed Model Fixed Effects σ 2 β β N(0, σ 2 βi) is not estimated, and is usually assumed to be large (or often in non-bayesian models) Random Effects σ 2 u is estimated. u N(0, σ 2 ui)
23 Linear Mixed Model > m2<-mcmcglmm(y ~ limit + year + day, random=~day, data=traffic)
24 Linear Mixed Model > m2<-mcmcglmm(y ~ limit + year + day, random=~day, data=traffic) > summary(m2) Iterations = 3001:12991 Thinning interval = 10 Sample size = 1000 DIC: G-structure: ~day post.mean l-95% CI u-95% CI eff.samp day R-structure: ~units post.mean l-95% CI u-95% CI eff.samp units Location effects: y ~ limit + year + day post.mean l-95% CI u-95% CI eff.samp pmcmc (Intercept) <0.001 *** limityes <0.001 *** year day Signif. codes: 0 '***' '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
25 Linear Mixed Model: Credible Intervals > plot(m2$vcv) Trace of day Density of day Iterations N = 1000 Bandwidth = Trace of units Density of units Iterations N = 1000 Bandwidth = Figure: Time-series of MCMC output (left) and smoothed posterior distribution (right) for the variance components.
26 Linear Mixed Model: Credible Intervals > plot(cbind(m2$vcv), type="l") units day Figure: MCMC trace through the joint posterior distribution for the two variance components.
27 Linear Mixed Model: Credible Intervals > r2<-m2$vcv[,"day"]/(m2$vcv[,"day"]+m2$vcv[,"units"]) Iterations N = 1000 Bandwidth = Figure: Time-series of MCMC output (left) and smoothed posterior distribution (right) for the proportion of variance explained by day.
28 Linear Model Diagnostics > hist(traffic$y-predict(m2)) Frequency Residual Figure: Histogram of residuals from model m1 which assumes they followed a Normal distribution.
29 Generalised Linear Model Link function g(): log log(e[y]) = Xβ
30 Generalised Linear Model Link function g(): log log(e[y]) = E[y] = E[y] = Xβ log 1 (Xβ) exp(xβ)
31 Generalised Linear Model Link function g(): log log(e[y]) = E[y] = E[y] = Xβ log 1 (Xβ) exp(xβ) Distribution: Poisson y Pois(λ = exp(xβ))
32 Generalised Linear Mixed Model: MCMCglmm A latent variable l where g 1 (l) is the distribution parameter:
33 Generalised Linear Mixed Model: MCMCglmm A latent variable l where g 1 (l) is the distribution parameter: y Pois(λ = exp(l))
34 Generalised Linear Mixed Model: MCMCglmm A latent variable l where g 1 (l) is the distribution parameter: y Pois(λ = exp(l)) then apply a standard linear model for the latent variables: l N(Wθ, Iσ 2 e )
35 Generalised Linear Mixed Model: MCMCglmm A latent variable l where g 1 (l) is the distribution parameter: y Pois(λ = exp(l)) then apply a standard linear model for the latent variables: l N(Wθ, Iσ 2 e ) Standard Poisson glm assumes σ 2 e = 0.
36 Generalised Linear Model: Poisson > m3<-mcmcglmm(y ~ limit + year + day, data=traffic, family="poisson")
37 Generalised Linear Model: Poisson > m3<-mcmcglmm(y ~ limit + year + day, data=traffic, family="poisson") > summary(m3) Iterations = 3001:12991 Thinning interval = 10 Sample size = 1000 DIC: R-structure: ~units post.mean l-95% CI u-95% CI eff.samp units Location effects: y ~ limit + year + day post.mean l-95% CI u-95% CI eff.samp pmcmc (Intercept) <0.001 *** limityes ** year day * --- Signif. codes: 0 '***' '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
38 Generalised Linear Model: Poisson > prior<-list(r=list(v=0.01, fix=1)) > m4<-mcmcglmm(y ~ limit + year + day, data=traffic, family="poisson", prior=prior)
39 Generalised Linear Model: Poisson > prior<-list(r=list(v=0.01, fix=1)) > m4<-mcmcglmm(y ~ limit + year + day, data=traffic, family="poisson", prior=prior) > summary(m4) Iterations = 3001:12991 Thinning interval = 10 Sample size = 1000 DIC: R-structure: ~units post.mean l-95% CI u-95% CI eff.samp units Location effects: y ~ limit + year + day post.mean l-95% CI u-95% CI eff.samp pmcmc (Intercept) <0.001 *** limityes <0.001 *** year day <0.001 *** --- Signif. codes: 0 '***' '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
40 Generalised Linear Mixed Model: Poisson > m5<-mcmcglmm(y ~ limit + year + day, random=~day, data=traffic, family="poisson")
41 Generalised Linear Mixed Model: Poisson > m5<-mcmcglmm(y ~ limit + year + day, random=~day, data=traffic, family="poisson") > summary(m5) Iterations = 3001:12991 Thinning interval = 10 Sample size = 1000 DIC: G-structure: ~day post.mean l-95% CI u-95% CI eff.samp day R-structure: ~units post.mean l-95% CI u-95% CI eff.samp units e Location effects: y ~ limit + year + day post.mean l-95% CI u-95% CI eff.samp pmcmc (Intercept) <0.001 *** limityes <0.001 *** year ** day Signif. codes: 0 '***' '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
42 Generalised Linear Mixed Model: Poisson Trace of day Density of day Iterations N = 1000 Bandwidth = Trace of units Density of units Iterations N = 1000 Bandwidth = 1.834e 05 Figure: Time-series of MCMC output (left) and smoothed posterior distribution (right) with flat (improper) priors on the variance components.
43 Generalised Linear Model > prior<-list(r=list(v=1, nu=0.002), G=list(G1=list(V=1, nu=1, alpha.mu=0, alpha.v=1000)))
44 Generalised Linear Model > prior<-list(r=list(v=1, nu=0.002), G=list(G1=list(V=1, nu=1, alpha.mu=0, alpha.v=1000))) > m5.b<-mcmcglmm(y ~ limit + year + day, random=~day, data=traffic, family="poisson", + prior=prior, nitt=13000*10, thin=10*10, burnin=3000*10) > plot(m5.b$vcv) Trace of day Density of day Iterations N = 1000 Bandwidth = Trace of units Density of units Iterations N = 1000 Bandwidth = Figure: Time-series of MCMC output (left) and smoothed posterior distribution (right) with proper priors on the variance components.
45 Distributions Binomial Multinomial Gaussian Poisson Ordinal Exponential Geometric Threshold Zero-inflated Poisson Zero-altered Poisson Hurdle Poisson Zero-inflated Binomial Censored Gaussian Censored Poisson Censored Exponential
46 Generalised Linear Mixed Model: Binary A latent variable l where g 1 (l) is the distribution parameter:
47 Generalised Linear Mixed Model: Binary A latent variable l where g 1 (l) is the distribution parameter: y Binom(Pr = probit 1 (l)) then apply a standard linear model for the latent variables: l N(Wθ, Iσ 2 e )
48 Generalised Linear Mixed Model: Binary > Traffic$y2<-as.numeric(Traffic$y>20) > m6<-mcmcglmm(y2 ~ limit + year + day, random=~day, data=traffic, + family="ordinal", slice=true)
49 Generalised Linear Mixed Model: Binary > Traffic$y2<-as.numeric(Traffic$y>20) > m6<-mcmcglmm(y2 ~ limit + year + day, random=~day, data=traffic, + family="ordinal", slice=true) > plot(m6$vcv) Trace of day Density of day 0e+00 2e+15 4e Iterations 0e+00 4e 15 0e+00 2e+15 4e+15 N = 1000 Bandwidth = 6.716e+13 Trace of units Density of units 0e+00 2e+15 4e Iterations 0e+00 3e 15 0e+00 2e+15 4e+15 N = 1000 Bandwidth = 8.95e+13 Figure: Time-series of MCMC output (left) and smoothed posterior distribution (right) for variance components.
50 Generalised Linear Mixed Model: Binary > prior<-list(r=list(v=1, fix=1), G=list(G1=list(V=1, nu=1, alpha.mu=0, alpha.v=1000))) > m7<-mcmcglmm(y2 ~ limit + year + day, random=~day, data=traffic, + family="ordinal", slice=true, prior=prior)
51 Generalised Linear Mixed Model: Binary > prior<-list(r=list(v=1, fix=1), G=list(G1=list(V=1, nu=1, alpha.mu=0, alpha.v=1000))) > m7<-mcmcglmm(y2 ~ limit + year + day, random=~day, data=traffic, + family="ordinal", slice=true, prior=prior) > plot(m7$vcv) Trace of day Density of day Iterations N = 1000 Bandwidth = Trace of units Density of units Iterations Figure: Time-series of MCMC output (left) and smoothed posterior distribution (right) for variance components.
52 Generalised Linear Mixed Model: Binary > prior<-list(r=list(v=0.5, fix=1), G=list(G1=list(V=1, nu=1, alpha.mu=0, alpha.v=1000))) > m8<-mcmcglmm(y2 ~ limit + year + day, random=~day, data=traffic, + family="ordinal", slice=true, prior=prior)
53 Generalised Linear Mixed Model: Binary > prior<-list(r=list(v=0.5, fix=1), G=list(G1=list(V=1, nu=1, alpha.mu=0, alpha.v=1000))) > m8<-mcmcglmm(y2 ~ limit + year + day, random=~day, data=traffic, + family="ordinal", slice=true, prior=prior) > plot(mcmc.list(m7$vcv, m8$vcv)) Trace of day Density of day Iterations N = 1000 Bandwidth = Trace of units Density of units Iterations N = 1000 Bandwidth = Figure: Time-series of MCMC output (left) and smoothed posterior distribution (right) for variance components with σ 2 e = 1 (black) and σ 2 e = 0.5 (red)
54 Generalised Linear Mixed Model: Binary > plot(mcmc.list(m7$sol[,"limityes"], m8$sol[,"limityes"])) Iterations N = 1000 Bandwidth = Figure: Time-series of MCMC output (left) and smoothed posterior distribution (right) for the effect of a speed limit with σ 2 e = 1 (black) and σ 2 e = 0.5 (red)
55 Generalised Linear Mixed Model: Binary > res.7<-m7$sol[,"limityes"]/sqrt(m7$vcv[,"units"]+1) > res.8<-m8$sol[,"limityes"]/sqrt(m8$vcv[,"units"]+1)
56 Generalised Linear Mixed Model: Binary > res.7<-m7$sol[,"limityes"]/sqrt(m7$vcv[,"units"]+1) > res.8<-m8$sol[,"limityes"]/sqrt(m8$vcv[,"units"]+1) > plot(mcmc.list(res.7, res.8)) Iterations N = 1000 Bandwidth = Figure: Time-series of MCMC output (left) and smoothed posterior distribution (right) for the scaled effect of a speed limit with σ 2 e = 1 (black) and σ 2 e = 0.5 (red)
57 Generalised Linear Mixed Model: Binary Pr = probit 1 (l) Wθ+e Liability
58 Generalised Linear Mixed Model: Binary Pr = probit 1 (l) Wθ+e Pr Liability
59 Generalised Linear Mixed Model: Binary Pr = probit 1 (l) ε Wθ+e Pr Liability
60 Generalised Linear Mixed Model: Binary Pr = probit 1 (l) ε e Wθ Pr Liability
61 Generalised Linear Mixed Model: Binary Pr = probit 1 (l) ε e Wθσ e ε Pr Liability
62 Generalised Linear Mixed Model: Binary > prior<-list(r=list(v=1, fix=1), G=list(G1=list(V=1, nu=1, alpha.mu=0, alpha.v=1000))) > m9<-mcmcglmm(y2 ~ limit + year + day, random=~day, data=traffic, + family="threshold", prior=prior)
63 Generalised Linear Mixed Model: Binary > prior<-list(r=list(v=1, fix=1), G=list(G1=list(V=1, nu=1, alpha.mu=0, alpha.v=1000))) > m9<-mcmcglmm(y2 ~ limit + year + day, random=~day, data=traffic, + family="threshold", prior=prior) > plot(mcmc.list(res.7, res.8, m9$sol[,"limityes"])) Iterations N = 1000 Bandwidth = Figure: Time-series of MCMC output (left) and smoothed posterior distribution (right) for the scaled effect of a speed limit with σe 2 = 1 (black), σe 2 = 0.5 (red) and σe 2 = 0 (green)
MCMCglmm Course Notes. Jarrod Hadfield
MCMCglmm Course Notes Jarrod Hadfield (j.hadfield@ed.ac.uk) March 17, 2014 Introduction These are (incomplete) course notes about generalised linear mixed models (GLMM). Special emphasis is placed on understanding
More informationMCMCglmm: Markov chain Monte Carlo methods for Generalised Linear Mixed Models.
MCMCglmm: Markov chain Monte Carlo methods for Generalised Linear Mixed Models. J. D. Hadfield February 10, 2010 1 Contents 1 An Empirical Example 3 1.1 Simple univariate model - Binomial response...........
More informationTwo Hours. Mathematical formula books and statistical tables are to be provided THE UNIVERSITY OF MANCHESTER. 26 May :00 16:00
Two Hours MATH38052 Mathematical formula books and statistical tables are to be provided THE UNIVERSITY OF MANCHESTER GENERALISED LINEAR MODELS 26 May 2016 14:00 16:00 Answer ALL TWO questions in Section
More informationLinear Regression Models P8111
Linear Regression Models P8111 Lecture 25 Jeff Goldsmith April 26, 2016 1 of 37 Today s Lecture Logistic regression / GLMs Model framework Interpretation Estimation 2 of 37 Linear regression Course started
More informationMultilevel Statistical Models: 3 rd edition, 2003 Contents
Multilevel Statistical Models: 3 rd edition, 2003 Contents Preface Acknowledgements Notation Two and three level models. A general classification notation and diagram Glossary Chapter 1 An introduction
More informationLecture 9 Multi-Trait Models, Binary and Count Traits
Lecture 9 Multi-Trait Models, Binary and Count Traits Guilherme J. M. Rosa University of Wisconsin-Madison Mixed Models in Quantitative Genetics SISG, Seattle 18 0 September 018 OUTLINE Multiple-trait
More informationA general mixed model approach for spatio-temporal regression data
A general mixed model approach for spatio-temporal regression data Thomas Kneib, Ludwig Fahrmeir & Stefan Lang Department of Statistics, Ludwig-Maximilians-University Munich 1. Spatio-temporal regression
More informationStatistics 203: Introduction to Regression and Analysis of Variance Course review
Statistics 203: Introduction to Regression and Analysis of Variance Course review Jonathan Taylor - p. 1/?? Today Review / overview of what we learned. - p. 2/?? General themes in regression models Specifying
More informationAdvanced Quantitative Methods: limited dependent variables
Advanced Quantitative Methods: Limited Dependent Variables I University College Dublin 2 April 2013 1 2 3 4 5 Outline Model Measurement levels 1 2 3 4 5 Components Model Measurement levels Two components
More informationCentre for Biodiversity Dynamics, Department of Biology, NTNU, NO-7491 Trondheim, Norway,
Animal models and Integrated Nested Laplace Approximations Anna Marie Holand,1, Ingelin Steinsland, Sara Martino,, Henrik Jensen Centre for Biodiversity Dynamics, Department of Biology, NTNU, NO-7491 Trondheim,
More informationAnalysing geoadditive regression data: a mixed model approach
Analysing geoadditive regression data: a mixed model approach Institut für Statistik, Ludwig-Maximilians-Universität München Joint work with Ludwig Fahrmeir & Stefan Lang 25.11.2005 Spatio-temporal regression
More informationBayesian linear regression
Bayesian linear regression Linear regression is the basis of most statistical modeling. The model is Y i = X T i β + ε i, where Y i is the continuous response X i = (X i1,..., X ip ) T is the corresponding
More informationLinear Regression. Data Model. β, σ 2. Process Model. ,V β. ,s 2. s 1. Parameter Model
Regression: Part II Linear Regression y~n X, 2 X Y Data Model β, σ 2 Process Model Β 0,V β s 1,s 2 Parameter Model Assumptions of Linear Model Homoskedasticity No error in X variables Error in Y variables
More informationStat 535 C - Statistical Computing & Monte Carlo Methods. Lecture 15-7th March Arnaud Doucet
Stat 535 C - Statistical Computing & Monte Carlo Methods Lecture 15-7th March 2006 Arnaud Doucet Email: arnaud@cs.ubc.ca 1 1.1 Outline Mixture and composition of kernels. Hybrid algorithms. Examples Overview
More informationGeneralized Linear Models for Non-Normal Data
Generalized Linear Models for Non-Normal Data Today s Class: 3 parts of a generalized model Models for binary outcomes Complications for generalized multivariate or multilevel models SPLH 861: Lecture
More informationSEEC Toolbox seminars
SEEC Toolbox seminars Data Exploration Greg Distiller 29th November Motivation The availability of sophisticated statistical tools has grown. But rubbish in - rubbish out.... even the well-known assumptions
More information13.1 Causal effects with continuous mediator and. predictors in their equations. The definitions for the direct, total indirect,
13 Appendix 13.1 Causal effects with continuous mediator and continuous outcome Consider the model of Section 3, y i = β 0 + β 1 m i + β 2 x i + β 3 x i m i + β 4 c i + ɛ 1i, (49) m i = γ 0 + γ 1 x i +
More informationDIC, AIC, BIC, PPL, MSPE Residuals Predictive residuals
DIC, AIC, BIC, PPL, MSPE Residuals Predictive residuals Overall Measures of GOF Deviance: this measures the overall likelihood of the model given a parameter vector D( θ) = 2 log L( θ) This can be averaged
More informationUsing the Delta Method to Construct Confidence Intervals for Predicted Probabilities, Rates, and Discrete Changes 1
Using the Delta Method to Construct Confidence Intervals for Predicted Probabilities, Rates, Discrete Changes 1 JunXuJ.ScottLong Indiana University 2005-02-03 1 General Formula The delta method is a general
More informationAMS-207: Bayesian Statistics
Linear Regression How does a quantity y, vary as a function of another quantity, or vector of quantities x? We are interested in p(y θ, x) under a model in which n observations (x i, y i ) are exchangeable.
More informationLecture 16: Mixtures of Generalized Linear Models
Lecture 16: Mixtures of Generalized Linear Models October 26, 2006 Setting Outline Often, a single GLM may be insufficiently flexible to characterize the data Setting Often, a single GLM may be insufficiently
More informationNELS 88. Latent Response Variable Formulation Versus Probability Curve Formulation
NELS 88 Table 2.3 Adjusted odds ratios of eighth-grade students in 988 performing below basic levels of reading and mathematics in 988 and dropping out of school, 988 to 990, by basic demographics Variable
More informationModelling geoadditive survival data
Modelling geoadditive survival data Thomas Kneib & Ludwig Fahrmeir Department of Statistics, Ludwig-Maximilians-University Munich 1. Leukemia survival data 2. Structured hazard regression 3. Mixed model
More informationFrailty Modeling for Spatially Correlated Survival Data, with Application to Infant Mortality in Minnesota By: Sudipto Banerjee, Mela. P.
Frailty Modeling for Spatially Correlated Survival Data, with Application to Infant Mortality in Minnesota By: Sudipto Banerjee, Melanie M. Wall, Bradley P. Carlin November 24, 2014 Outlines of the talk
More informationgeorglm : a package for generalised linear spatial models introductory session
georglm : a package for generalised linear spatial models introductory session Ole F. Christensen & Paulo J. Ribeiro Jr. Last update: October 18, 2017 The objective of this page is to introduce the reader
More informationChecking the Poisson assumption in the Poisson generalized linear model
Checking the Poisson assumption in the Poisson generalized linear model The Poisson regression model is a generalized linear model (glm) satisfying the following assumptions: The responses y i are independent
More informationChapter 4 Multi-factor Treatment Designs with Multiple Error Terms 93
Contents Preface ix Chapter 1 Introduction 1 1.1 Types of Models That Produce Data 1 1.2 Statistical Models 2 1.3 Fixed and Random Effects 4 1.4 Mixed Models 6 1.5 Typical Studies and the Modeling Issues
More informationGeneralized Linear. Mixed Models. Methods and Applications. Modern Concepts, Walter W. Stroup. Texts in Statistical Science.
Texts in Statistical Science Generalized Linear Mixed Models Modern Concepts, Methods and Applications Walter W. Stroup CRC Press Taylor & Francis Croup Boca Raton London New York CRC Press is an imprint
More informationRecent advances in statistical methods for DNA-based prediction of complex traits
Recent advances in statistical methods for DNA-based prediction of complex traits Mintu Nath Biomathematics & Statistics Scotland, Edinburgh 1 Outline Background Population genetics Animal model Methodology
More informationIntroduction to GSEM in Stata
Introduction to GSEM in Stata Christopher F Baum ECON 8823: Applied Econometrics Boston College, Spring 2016 Christopher F Baum (BC / DIW) Introduction to GSEM in Stata Boston College, Spring 2016 1 /
More informationMachine learning: Hypothesis testing. Anders Hildeman
Location of trees 0 Observed trees 50 100 150 200 250 300 350 400 450 500 0 100 200 300 400 500 600 700 800 900 1000 Figur: Observed points pattern of the tree specie Beilschmiedia pendula. Location of
More informationGeneralized Linear Models
York SPIDA John Fox Notes Generalized Linear Models Copyright 2010 by John Fox Generalized Linear Models 1 1. Topics I The structure of generalized linear models I Poisson and other generalized linear
More informationBayesian Inference on Joint Mixture Models for Survival-Longitudinal Data with Multiple Features. Yangxin Huang
Bayesian Inference on Joint Mixture Models for Survival-Longitudinal Data with Multiple Features Yangxin Huang Department of Epidemiology and Biostatistics, COPH, USF, Tampa, FL yhuang@health.usf.edu January
More informationReview. Timothy Hanson. Department of Statistics, University of South Carolina. Stat 770: Categorical Data Analysis
Review Timothy Hanson Department of Statistics, University of South Carolina Stat 770: Categorical Data Analysis 1 / 22 Chapter 1: background Nominal, ordinal, interval data. Distributions: Poisson, binomial,
More informationChapter 1. Modeling Basics
Chapter 1. Modeling Basics What is a model? Model equation and probability distribution Types of model effects Writing models in matrix form Summary 1 What is a statistical model? A model is a mathematical
More informationLecture 14: Introduction to Poisson Regression
Lecture 14: Introduction to Poisson Regression Ani Manichaikul amanicha@jhsph.edu 8 May 2007 1 / 52 Overview Modelling counts Contingency tables Poisson regression models 2 / 52 Modelling counts I Why
More informationModelling counts. Lecture 14: Introduction to Poisson Regression. Overview
Modelling counts I Lecture 14: Introduction to Poisson Regression Ani Manichaikul amanicha@jhsph.edu Why count data? Number of traffic accidents per day Mortality counts in a given neighborhood, per week
More informationRegression models. Generalized linear models in R. Normal regression models are not always appropriate. Generalized linear models. Examples.
Regression models Generalized linear models in R Dr Peter K Dunn http://www.usq.edu.au Department of Mathematics and Computing University of Southern Queensland ASC, July 00 The usual linear regression
More informationContents. Part I: Fundamentals of Bayesian Inference 1
Contents Preface xiii Part I: Fundamentals of Bayesian Inference 1 1 Probability and inference 3 1.1 The three steps of Bayesian data analysis 3 1.2 General notation for statistical inference 4 1.3 Bayesian
More informationBayesian Linear Regression
Bayesian Linear Regression Sudipto Banerjee 1 Biostatistics, School of Public Health, University of Minnesota, Minneapolis, Minnesota, U.S.A. September 15, 2010 1 Linear regression models: a Bayesian perspective
More informationComparing IRT with Other Models
Comparing IRT with Other Models Lecture #14 ICPSR Item Response Theory Workshop Lecture #14: 1of 45 Lecture Overview The final set of slides will describe a parallel between IRT and another commonly used
More informationMultiple regression: Categorical dependent variables
Multiple : Categorical Johan A. Elkink School of Politics & International Relations University College Dublin 28 November 2016 1 2 3 4 Outline 1 2 3 4 models models have a variable consisting of two categories.
More informationNon-Gaussian Response Variables
Non-Gaussian Response Variables What is the Generalized Model Doing? The fixed effects are like the factors in a traditional analysis of variance or linear model The random effects are different A generalized
More informationThe linear model is the most fundamental of all serious statistical models encompassing:
Linear Regression Models: A Bayesian perspective Ingredients of a linear model include an n 1 response vector y = (y 1,..., y n ) T and an n p design matrix (e.g. including regressors) X = [x 1,..., x
More informationPATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 2: PROBABILITY DISTRIBUTIONS
PATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 2: PROBABILITY DISTRIBUTIONS Parametric Distributions Basic building blocks: Need to determine given Representation: or? Recall Curve Fitting Binary Variables
More informationSingle-level Models for Binary Responses
Single-level Models for Binary Responses Distribution of Binary Data y i response for individual i (i = 1,..., n), coded 0 or 1 Denote by r the number in the sample with y = 1 Mean and variance E(y) =
More informationGeneralised linear models. Response variable can take a number of different formats
Generalised linear models Response variable can take a number of different formats Structure Limitations of linear models and GLM theory GLM for count data GLM for presence \ absence data GLM for proportion
More informationWU Weiterbildung. Linear Mixed Models
Linear Mixed Effects Models WU Weiterbildung SLIDE 1 Outline 1 Estimation: ML vs. REML 2 Special Models On Two Levels Mixed ANOVA Or Random ANOVA Random Intercept Model Random Coefficients Model Intercept-and-Slopes-as-Outcomes
More informationSemiparametric Generalized Linear Models
Semiparametric Generalized Linear Models North American Stata Users Group Meeting Chicago, Illinois Paul Rathouz Department of Health Studies University of Chicago prathouz@uchicago.edu Liping Gao MS Student
More informationNow consider the case where E(Y) = µ = Xβ and V (Y) = σ 2 G, where G is diagonal, but unknown.
Weighting We have seen that if E(Y) = Xβ and V (Y) = σ 2 G, where G is known, the model can be rewritten as a linear model. This is known as generalized least squares or, if G is diagonal, with trace(g)
More informationA Fully Nonparametric Modeling Approach to. BNP Binary Regression
A Fully Nonparametric Modeling Approach to Binary Regression Maria Department of Applied Mathematics and Statistics University of California, Santa Cruz SBIES, April 27-28, 2012 Outline 1 2 3 Simulation
More informationGeneralized linear models
Generalized linear models Christopher F Baum ECON 8823: Applied Econometrics Boston College, Spring 2016 Christopher F Baum (BC / DIW) Generalized linear models Boston College, Spring 2016 1 / 1 Introduction
More informationLatent Variable Models for Binary Data. Suppose that for a given vector of explanatory variables x, the latent
Latent Variable Models for Binary Data Suppose that for a given vector of explanatory variables x, the latent variable, U, has a continuous cumulative distribution function F (u; x) and that the binary
More informationTento projekt je spolufinancován Evropským sociálním fondem a Státním rozpočtem ČR InoBio CZ.1.07/2.2.00/
Tento projekt je spolufinancován Evropským sociálním fondem a Státním rozpočtem ČR InoBio CZ.1.07/2.2.00/28.0018 Statistical Analysis in Ecology using R Linear Models/GLM Ing. Daniel Volařík, Ph.D. 13.
More informationPackage effectfusion
Package November 29, 2016 Title Bayesian Effect Fusion for Categorical Predictors Version 1.0 Date 2016-11-21 Author Daniela Pauger [aut, cre], Helga Wagner [aut], Gertraud Malsiner-Walli [aut] Maintainer
More informationSubject CS1 Actuarial Statistics 1 Core Principles
Institute of Actuaries of India Subject CS1 Actuarial Statistics 1 Core Principles For 2019 Examinations Aim The aim of the Actuarial Statistics 1 subject is to provide a grounding in mathematical and
More informationStat 542: Item Response Theory Modeling Using The Extended Rank Likelihood
Stat 542: Item Response Theory Modeling Using The Extended Rank Likelihood Jonathan Gruhl March 18, 2010 1 Introduction Researchers commonly apply item response theory (IRT) models to binary and ordinal
More informationSTA 216, GLM, Lecture 16. October 29, 2007
STA 216, GLM, Lecture 16 October 29, 2007 Efficient Posterior Computation in Factor Models Underlying Normal Models Generalized Latent Trait Models Formulation Genetic Epidemiology Illustration Structural
More informationStat 5101 Lecture Notes
Stat 5101 Lecture Notes Charles J. Geyer Copyright 1998, 1999, 2000, 2001 by Charles J. Geyer May 7, 2001 ii Stat 5101 (Geyer) Course Notes Contents 1 Random Variables and Change of Variables 1 1.1 Random
More informationGibbs Sampling in Latent Variable Models #1
Gibbs Sampling in Latent Variable Models #1 Econ 690 Purdue University Outline 1 Data augmentation 2 Probit Model Probit Application A Panel Probit Panel Probit 3 The Tobit Model Example: Female Labor
More informationPackage HGLMMM for Hierarchical Generalized Linear Models
Package HGLMMM for Hierarchical Generalized Linear Models Marek Molas Emmanuel Lesaffre Erasmus MC Erasmus Universiteit - Rotterdam The Netherlands ERASMUSMC - Biostatistics 20-04-2010 1 / 52 Outline General
More informationMULTILEVEL IMPUTATION 1
MULTILEVEL IMPUTATION 1 Supplement B: MCMC Sampling Steps and Distributions for Two-Level Imputation This document gives technical details of the full conditional distributions used to draw regression
More informationAddition to PGLR Chap 6
Arizona State University From the SelectedWorks of Joseph M Hilbe August 27, 216 Addition to PGLR Chap 6 Joseph M Hilbe, Arizona State University Available at: https://works.bepress.com/joseph_hilbe/69/
More informationNon-maximum likelihood estimation and statistical inference for linear and nonlinear mixed models
Optimum Design for Mixed Effects Non-Linear and generalized Linear Models Cambridge, August 9-12, 2011 Non-maximum likelihood estimation and statistical inference for linear and nonlinear mixed models
More informationA strategy for modelling count data which may have extra zeros
A strategy for modelling count data which may have extra zeros Alan Welsh Centre for Mathematics and its Applications Australian National University The Data Response is the number of Leadbeater s possum
More informationPreface Introduction to Statistics and Data Analysis Overview: Statistical Inference, Samples, Populations, and Experimental Design The Role of
Preface Introduction to Statistics and Data Analysis Overview: Statistical Inference, Samples, Populations, and Experimental Design The Role of Probability Sampling Procedures Collection of Data Measures
More informationDistribution Assumptions
Merlise Clyde Duke University November 22, 2016 Outline Topics Normality & Transformations Box-Cox Nonlinear Regression Readings: Christensen Chapter 13 & Wakefield Chapter 6 Linear Model Linear Model
More informationPart 8: GLMs and Hierarchical LMs and GLMs
Part 8: GLMs and Hierarchical LMs and GLMs 1 Example: Song sparrow reproductive success Arcese et al., (1992) provide data on a sample from a population of 52 female song sparrows studied over the course
More information3 Joint Distributions 71
2.2.3 The Normal Distribution 54 2.2.4 The Beta Density 58 2.3 Functions of a Random Variable 58 2.4 Concluding Remarks 64 2.5 Problems 64 3 Joint Distributions 71 3.1 Introduction 71 3.2 Discrete Random
More informationSTAT5044: Regression and Anova
STAT5044: Regression and Anova Inyoung Kim 1 / 18 Outline 1 Logistic regression for Binary data 2 Poisson regression for Count data 2 / 18 GLM Let Y denote a binary response variable. Each observation
More information36-463/663: Multilevel & Hierarchical Models
36-463/663: Multilevel & Hierarchical Models (P)review: in-class midterm Brian Junker 132E Baker Hall brian@stat.cmu.edu 1 In-class midterm Closed book, closed notes, closed electronics (otherwise I have
More informationExploring Hierarchical Linear Mixed Models
Exploring Hierarchical Linear Mixed Models 1/49 Last time... A Greenhouse Experiment testing C:N Ratios Sam was testing how changing the C:N Ratio of soil affected plant leaf growth. He had 3 treatments.
More informationCommunity Health Needs Assessment through Spatial Regression Modeling
Community Health Needs Assessment through Spatial Regression Modeling Glen D. Johnson, PhD CUNY School of Public Health glen.johnson@lehman.cuny.edu Objectives: Assess community needs with respect to particular
More informationBayesian model selection for computer model validation via mixture model estimation
Bayesian model selection for computer model validation via mixture model estimation Kaniav Kamary ATER, CNAM Joint work with É. Parent, P. Barbillon, M. Keller and N. Bousquet Outline Computer model validation
More informationLecture 9 STK3100/4100
Lecture 9 STK3100/4100 27. October 2014 Plan for lecture: 1. Linear mixed models cont. Models accounting for time dependencies (Ch. 6.1) 2. Generalized linear mixed models (GLMM, Ch. 13.1-13.3) Examples
More informationTobit and Selection Models
Tobit and Selection Models Class Notes Manuel Arellano November 24, 2008 Censored Regression Illustration : Top-coding in wages Suppose Y log wages) are subject to top coding as is often the case with
More informationFast Likelihood-Free Inference via Bayesian Optimization
Fast Likelihood-Free Inference via Bayesian Optimization Michael Gutmann https://sites.google.com/site/michaelgutmann University of Helsinki Aalto University Helsinki Institute for Information Technology
More informationSCHOOL OF MATHEMATICS AND STATISTICS. Linear and Generalised Linear Models
SCHOOL OF MATHEMATICS AND STATISTICS Linear and Generalised Linear Models Autumn Semester 2017 18 2 hours Attempt all the questions. The allocation of marks is shown in brackets. RESTRICTED OPEN BOOK EXAMINATION
More informationHypothesis Testing. Econ 690. Purdue University. Justin L. Tobias (Purdue) Testing 1 / 33
Hypothesis Testing Econ 690 Purdue University Justin L. Tobias (Purdue) Testing 1 / 33 Outline 1 Basic Testing Framework 2 Testing with HPD intervals 3 Example 4 Savage Dickey Density Ratio 5 Bartlett
More informationRonald Christensen. University of New Mexico. Albuquerque, New Mexico. Wesley Johnson. University of California, Irvine. Irvine, California
Texts in Statistical Science Bayesian Ideas and Data Analysis An Introduction for Scientists and Statisticians Ronald Christensen University of New Mexico Albuquerque, New Mexico Wesley Johnson University
More informationStat 5102 Final Exam May 14, 2015
Stat 5102 Final Exam May 14, 2015 Name Student ID The exam is closed book and closed notes. You may use three 8 1 11 2 sheets of paper with formulas, etc. You may also use the handouts on brand name distributions
More informationHeriot-Watt University
Heriot-Watt University Heriot-Watt University Research Gateway Prediction of settlement delay in critical illness insurance claims by using the generalized beta of the second kind distribution Dodd, Erengul;
More informationGeneralized linear models
Generalized linear models Douglas Bates November 01, 2010 Contents 1 Definition 1 2 Links 2 3 Estimating parameters 5 4 Example 6 5 Model building 8 6 Conclusions 8 7 Summary 9 1 Generalized Linear Models
More informationIntroduction to the Generalized Linear Model: Logistic regression and Poisson regression
Introduction to the Generalized Linear Model: Logistic regression and Poisson regression Statistical modelling: Theory and practice Gilles Guillot gigu@dtu.dk November 4, 2013 Gilles Guillot (gigu@dtu.dk)
More informationBayesian Inference in GLMs. Frequentists typically base inferences on MLEs, asymptotic confidence
Bayesian Inference in GLMs Frequentists typically base inferences on MLEs, asymptotic confidence limits, and log-likelihood ratio tests Bayesians base inferences on the posterior distribution of the unknowns
More informationLogistic regression. 11 Nov Logistic regression (EPFL) Applied Statistics 11 Nov / 20
Logistic regression 11 Nov 2010 Logistic regression (EPFL) Applied Statistics 11 Nov 2010 1 / 20 Modeling overview Want to capture important features of the relationship between a (set of) variable(s)
More informationChapter 4 - Fundamentals of spatial processes Lecture notes
Chapter 4 - Fundamentals of spatial processes Lecture notes Geir Storvik January 21, 2013 STK4150 - Intro 2 Spatial processes Typically correlation between nearby sites Mostly positive correlation Negative
More informationEfficient Bayesian Multivariate Surface Regression
Efficient Bayesian Multivariate Surface Regression Feng Li (joint with Mattias Villani) Department of Statistics, Stockholm University October, 211 Outline of the talk 1 Flexible regression models 2 The
More informationAccounting for Calibration Uncertainty in Spectral Analysis. David A. van Dyk
Bayesian Analysis of Accounting for in Spectral Analysis David A van Dyk Statistics Section, Imperial College London Joint work with Vinay Kashyap, Jin Xu, Alanna Connors, and Aneta Siegminowska ISIS May
More informationSupplementary File 5: Tutorial for MCMCglmm version. Tutorial 1 (MCMCglmm) - Estimating the heritability of birth weight
Supplementary File 5: Tutorial for MCMCglmm version Tutorial 1 (MCMCglmm) - Estimating the heritability of birth weight This tutorial will demonstrate how to run a univariate animal model using the R package
More informationGeneralized Linear Models. stat 557 Heike Hofmann
Generalized Linear Models stat 557 Heike Hofmann Outline Intro to GLM Exponential Family Likelihood Equations GLM for Binomial Response Generalized Linear Models Three components: random, systematic, link
More informationModeling Longitudinal Count Data with Excess Zeros and Time-Dependent Covariates: Application to Drug Use
Modeling Longitudinal Count Data with Excess Zeros and : Application to Drug Use University of Northern Colorado November 17, 2014 Presentation Outline I and Data Issues II Correlated Count Regression
More informationLogistic Regression in R. by Kerry Machemer 12/04/2015
Logistic Regression in R by Kerry Machemer 12/04/2015 Linear Regression {y i, x i1,, x ip } Linear Regression y i = dependent variable & x i = independent variable(s) y i = α + β 1 x i1 + + β p x ip +
More informationDiscrete Choice Modeling
[Part 6] 1/55 0 Introduction 1 Summary 2 Binary Choice 3 Panel Data 4 Bivariate Probit 5 Ordered Choice 6 7 Multinomial Choice 8 Nested Logit 9 Heterogeneity 10 Latent Class 11 Mixed Logit 12 Stated Preference
More informationRegression Model Building
Regression Model Building Setting: Possibly a large set of predictor variables (including interactions). Goal: Fit a parsimonious model that explains variation in Y with a small set of predictors Automated
More informationBayesian inference - Practical exercises Guiding document
Bayesian inference - Practical exercises Guiding document Elise Billoir, Marie Laure Delignette-Muller and Sandrine Charles ebilloir@pole-ecotox.fr marielaure.delignettemuller@vetagro-sup.fr sandrine.charles@univ-lyon1.fr
More informationDefault Priors and Effcient Posterior Computation in Bayesian
Default Priors and Effcient Posterior Computation in Bayesian Factor Analysis January 16, 2010 Presented by Eric Wang, Duke University Background and Motivation A Brief Review of Parameter Expansion Literature
More informationBayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes
Bayesian dynamic modeling for large space-time weather datasets using Gaussian predictive processes Andrew O. Finley 1 and Sudipto Banerjee 2 1 Department of Forestry & Department of Geography, Michigan
More informationGeneralized Linear Models
Generalized Linear Models Assumptions of Linear Model Homoskedasticity Model variance No error in X variables Errors in variables No missing data Missing data model Normally distributed error Error in
More informationReview: what is a linear model. Y = β 0 + β 1 X 1 + β 2 X 2 + A model of the following form:
Outline for today What is a generalized linear model Linear predictors and link functions Example: fit a constant (the proportion) Analysis of deviance table Example: fit dose-response data using logistic
More information