https://nd.psychstat.org/ 2026-04-20T15:34:35+00:00 https://nd.psychstat.org/ https://nd.psychstat.org/_media/wiki/dokuwiki.svg text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/blog text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/bootmedem V1 on Nov 22, 2008 V1 on Nov 23, 2008 text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/bootstrap_ci ## EM for the estimate the covariance matrix ## function to calculate the parameters par.est<-function(pM, pS){ sX2<-pS[1,1] sM2<-pS[2,2] sY2<-pS[3,3] sXM<-pS[1,2] sXY<-pS[1,3] sMY<-pS[2,3] mX <-pM[1] mM <-pM[2] mY <-pM[3] pa<-sXM/sX2 pb<-(sMY*sX2-sXM*sXY)/(sX2*sM2-sXM^2) pc<-(sXY*sM2-sXM*sMY)/(sX2*sM2-sXM^2) ps1<-sX2 ps2<-sM2-sXM^2/sX2 ps3<-(sX2*sM2*sY2-sX2*sMY^2-sM2*sXY^2-sY2*sXM^2+2*sXM*sXY*sMY)/(sX2*sM2-sXM^2) pm1<-mX pm2<-mM - pa*mX pm3<-mY - pb*mM - p… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/bootstrap_em_for_mediation_model bootstrap CI old 1 final for simulation Process results Data text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/data Data set 1 Age -- Reasoning -- EPT 89 14 99999 78 15 99999 77 31 99999 93 99999 99999 79 99999 99999 80 99999 3 82 99999 4 76 99999 7 68 99999 7 85 99999 9 77 99999 9 85 99999 11 73 99999 12 85 99999 13 71 99999 13 94 99999 15 85 99999 17 69 16 26 69 30 23 79 40 14 73 15 13 78 21 16 69 27 13 72 24 22 81 20 13 79 6 17 78 17 23 75 17 14 84 39 23 76 24 27 71 27 20 79 12 13 66 10 9 73 35 27 68 32 22 81 14 18 81 7 15 86 22 22 86 21 16 72 11 20 78 13 11 65 32 25 77 19 18 72 25 15 71 45 28 74 44 23 … text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/download_bootmedem_and_document BMEM Download The program can be downloaded here [download]. The BMEM manual can be download here [Manual]. Document Background The program deals with a mediation model specified as <jsmath>M = iM + aX + e_1</jsmath> <jsmath>Y = iY + bM + cX + e_2</jsmath> where <jsm>X</jsm>, <jsm>M</jsm>, and <jsm>Y</jsm> are input, mediator, and output variables. text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/em_fiml ML methods are based on the assumption that the missingness mechanism is at least MAR, but not necessarily MCAR; on the other hand, it is assumed that the data are normally distributed. Non-Bayesian methods--maximum likelihood (ML) approach with the EM algorithm The formulation of the EM algorithm made it feasible to compute ML estimates in many missing-data problems. EM algorithm may be used for maximum likelihood estimation. The EM algorithm is a method for obtaining maximum likelihood estimat… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/em_for_mediation_for_monents ## EM for the estimate the covariance matrix ## function to calculate the parameters par.est<-function(pM, pS){ sX2<-pS[1,1] sM2<-pS[2,2] sY2<-pS[3,3] sXM<-pS[1,2] sXY<-pS[1,3] sMY<-pS[2,3] mX <-pM[1] mM <-pM[2] mY <-pM[3] pa<-sXM/sX2 pb<-(sMY*sX2-sXM*sXY)/(sX2*sM2-sXM^2) pc<-(sXY*sM2-sXM*sMY)/(sX2*sM2-sXM^2) ps1<-sX2 ps2<-sM2-sXM^2/sX2 ps3<-(sX2*sM2*sY2-sX2*sMY^2-sM2*sXY^2-sY2*sXM^2+2*sXM*sXY*sMY)/(sX2*sM2-sXM^2) pm1<-mX pm2<-mM - pa*mX pm3<-mY - pb*mM - p… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/em_for_mediation_model_with_means ## EM for the mediation model ## function to calculate the parameters est<-function(S){ sX2<-S[1] sM2<-S[2] sY2<-S[3] sXM<-S[4] sXY<-S[5] sMY<-S[6] mX <-S[7] mM <-S[8] mY <-S[9] pa<-sXM/sX2 pb<-(sMY*sX2-sXM*sXY)/(sX2*sM2-sXM^2) pc<-(sXY*sM2-sXM*sMY)/(sX2*sM2-sXM^2) ps1<-sX2 ps2<-sM2-sXM^2/sX2 ps3<-(sX2*sM2*sY2-sX2*sMY^2-sM2*sXY^2-sY2*sXM^2+2*sXM*sXY*sMY)/(sX2*sM2-sXM^2) pm1<-mX pm2<-mM - pa*mX pm3<-mY - pb*mM - pc*mX return(c(pa,pb,pc,ps1,ps2,ps3,pm1,p… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/em_for_mediation_model ## EM for the mediation model ## function to calculate the parameters est<-function(S){ sX2<-S[1] sM2<-S[2] sY2<-S[3] sXM<-S[4] sXY<-S[5] sMY<-S[6] pa<-sXM/sX2 pb<-(sMY*sX2-sXM*sXY)/(sX2*sM2-sXM^2) pc<-(sXY*sM2-sXM*sMY)/(sX2*sM2-sXM^2) ps1<-sX2 ps2<-sM2-sXM^2/sX2 ps3<-(sX2*sM2*sY2-sX2*sMY^2-sM2*sXY^2-sY2*sXM^2+2*sXM*sXY*sMY)/(sX2*sM2-sXM^2) return(c(pa,pb,pc,ps1,ps2,ps3)) } ## expectation step EMe<-function(data, para){ X<-data[,1] Y<-data[,3] M<-data[,2] p… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/em_for_the_simple_regression ## EM for simple regression ## data x<-rnorm(100) y<-x+.5*rnorm(100) summary(lm(y~x-1)) y[91:100]<-NA a0<-1 s0<-.5 e<-1 while (e>.00001){ sx<-sum(x^2) sy<-sum(y[1:90]^2)+a0^2*sum(x[91:100]^2)+10*s0 sxy<-sum(x[1:90]*y[1:90])+a0*sum(x[91:100]^2) a<-sxy/sx s<-(sx*sy-sxy^2)/(sx^2) e<-(a0-a)^2+(s0-s)^2 a0<-a s0<-s } text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/exploratory_data_analysis <http://duncanjg.wordpress.com/2008/11/21/draft-beginners-r-course/> text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/final_for_simulation ## EM for the estimate the covariance matrix ## function to calculate the parameters par.est<-function(pM, pS){ sX2<-pS[1,1] sM2<-pS[2,2] sY2<-pS[3,3] sXM<-pS[1,2] sXY<-pS[1,3] sMY<-pS[2,3] mX <-pM[1] mM <-pM[2] mY <-pM[3] pa<-sXM/sX2 pb<-(sMY*sX2-sXM*sXY)/(sX2*sM2-sXM^2) pc<-(sXY*sM2-sXM*sMY)/(sX2*sM2-sXM^2) ps1<-sX2 ps2<-sM2-sXM^2/sX2 ps3<-(sX2*sM2*sY2-sX2*sMY^2-sM2*sXY^2-sY2*sXM^2+2*sXM*sXY*sMY)/(sX2*sM2-sXM^2) pm1<-mX pm2<-mM - pa*mX pm3<-mY - pb*mM - p… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/higer--order_invariance Programs Dec 17, 2008 R codes to process the results from Mplus for Table 1 & 2 R codes to process the results from Mplus for Table 3 R codes for simulation study in Experiment 4 text/html 2020-06-07T18:15:53+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/index Lab meetings Meeting schedules People * Johnny Zhang * Peggy Wang * Xin Tong * Haiyan Liu * Qian Zhang * Han Du * Charlotte Cunningham Classes * Advanced stats On-going projects * Bayesian differential equations * Bayesian differential equations by Johnny, Xin, Laura * Growth curve modeling with MNAR data by Johnny, Xin * Bayesian analysis using SAS by Johnny * Bayesian latent change score model by Johnny, Kevin Grimm * Using Wishart prior distribution in Bayes… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/mediation_model_maple_codes Iden := Matrix(3, 3, {(1, 1) = 1, (1, 2) = 0, (1, 3) = 0, (2, 1) = 0, (2, 2) = 1, (2, 3) = 0, (3, 1) = 0, (3, 2) = 0, (3, 3) = 1}) A := Matrix(3, 3, {(1, 1) = 0, (1, 2) = 0, (1, 3) = 0, (2, 1) = a, (2, 2) = 0, (2, 3) = 0, (3, 1) = c, (3, 2) = b, (3, 3) = 0}) S := Matrix(3, 3, {(1, 1) = sigma[1]^2, (1, 2) = 0, (1, 3) = 0, (2, 1) = 0, (2, 2) = sigma[2]^2, (2, 3) = 0, (3, 1) = 0, (3, 2) = 0, (3, 3) = sigma[3]^2}) with(linalg): omega := multiply(multiply(inverse(Iden-A), S), transpose(inverse(Id… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/nonparametric_bayesian STAT 882: Nonparametric Bayesian Inference Winter Quarter 2009 ---------- ^Instructors:| |Steve MacEachern. | |office Hours: Friday 9:30 - 10:30am in 205C Cockins Hall, and by appointment| Xinyi Xu. office Hours: Tuesday 9:30 - 10:30am in 440G Cockins Hall, and by appointment ^Lecture Hours & Location: text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/note_on_using_visual_c_2008 To add the library Go to Project -- Properties -- Configuration -- Linker -- Input -- Type in Additional dependencies (the lib names: newran03.lib) Configuration -- Linker -- general -- Additional library directories -- type in the path to the library text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/old_1 ## EM for the estimate the covariance matrix ## function to calculate the parameters par.est<-function(pM, pS){ sX2<-pS[1,1] sM2<-pS[2,2] sY2<-pS[3,3] sXM<-pS[1,2] sXY<-pS[1,3] sMY<-pS[2,3] mX <-pM[1] mM <-pM[2] mY <-pM[3] pa<-sXM/sX2 pb<-(sMY*sX2-sXM*sXY)/(sX2*sM2-sXM^2) pc<-(sXY*sM2-sXM*sMY)/(sX2*sM2-sXM^2) ps1<-sX2 ps2<-sM2-sXM^2/sX2 ps3<-(sX2*sM2*sY2-sX2*sMY^2-sM2*sXY^2-sY2*sXM^2+2*sXM*sXY*sMY)/(sX2*sM2-sXM^2) pm1<-mX pm2<-mM - pa*mX pm3<-mY - pb*mM - p… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/proc_mi data FitMiss; input Oxygen RunTime RunPulse @@; datalines; 44.609 11.37 178 45.313 10.07 185 54.297 8.65 156 59.571 . . 49.874 9.22 . 44.811 11.63 176 . 11.95 176 . 10.85 . 39.442 13.08 174 60.055 8.63 170 50.541 . . 37.388 14.03 186 44.754 11.12 176 47.273 . . 51.855 10.33 166 49.156 8.95 180 40.836 10.95 168 46.672 10.00 . 46.774 1… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/process_results ## Process the results from simulation for missing data mediation analysis setwd("H:\\Private\\bootmedem\\l-20") res<-read.table("res-N100-miss0.2.txt") sumres<-cbind(apply(res,2,mean),apply(res,2,sd)) ## estimates and the standard deviation cbind(sumres[37:41,1],sumres[77:81,1],sumres[117:121,1],sumres[157:161,1]) cbind(sumres[37:41,2],sumres[77:81,2],sumres[117:121,2],sumres[157:161,2]) ## coverage probability cvg<-NULL for (i in 0:3){ cvg<-cbind(cvg,c(sum(res[,38+40*i]<.3481 & res[,41+40… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/r_codes_for_simulation_study_in_experiment_4 ## Dec 14, 2008 ## Estimating the parameters est.g34.inv<-function(par, data, cor=TRUE){ x1<-data$x1 x2<-data$x2 ## factor correlation matrix phi<-matrix(c(1, par[1], par[2], par[1], 1, par[3], par[2], par[3], 1 ), nrow=3, byrow=T) ## factor residual variance psi1<-diag(1,9,9) for (i in 1:9){ psi1[i,i]<-par[3+i] } ## factor loading matrix lam11<-diag(1,3,3) for (i in 1:3){ lam11[i,i]<-par[12+i] } lam12<-matrix(par[16:33],nrow=6,byrow=T) lambda… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/r_codes_to_process_the_results_from_mplus_for_table_1_2 ## model 2 res<-scan("1g-2-500.txt") res.mat<-matrix(res, byrow=T, ncol=66) res.f<-res.mat[1:1000,26:28] cbind(apply(res.f,2,mean),apply(res.f,2,var)) ## for model 3 to 6 res<-scan("1g-5-500.txt") res.mat<-matrix(res, byrow=T, ncol=96) res.f<-res.mat[1:1000,41:43] cbind(apply(res.f,2,mean),apply(res.f,2,var)) ## for model 7 res<-scan("1g-7-500.txt") res.mat<-matrix(res, byrow=T, ncol=126) res.f<-res.mat[1:1000,56:58] cbind(apply(res.f,2,mean),apply(res.f,2,var)) text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/r_codes_to_process_the_results_from_mplus_for_table_3 ## handle the model comparison results from mplus v500<-scan('all_results500v.txt') v500m<-matrix(v500, nrow=1500, byrow=T) inv500<-scan('all_results500inv.txt') inv500m<-matrix(inv500, nrow=1497, byrow=T) v500f<-v500m[c(1:523, 525:1034, 1036:1279, 1281:1500), c(40:42, 169)] inv500f<-inv500m[, c(40:42, 163)] mean(v500f[,4]) var(v500f[,4]) ks.test(v500f[,4],'pchisq',23) mean(inv500f[,4]) var(inv500f[,4]) ks.test(inv500f[,4],'pchisq',27) chi.diff<-inv500f[,4]-v500f[,4] mean(chi.diff) var(c… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/task text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/test_winbugs test text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/use_of_server Use of Wiki Create a task * click the edit page link on the bottom * Input code like <todo>Task to be done</todo> <todo>Task to be done</todo> * You may add time as a section title as ==== Thursday ==== <todo>Task to be done</todo> Thursday text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/useful_things * features about multivariate normal distribution [multivariate normal] text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/using_winbugs_on_the_notre_dame_clusters Steps * Download OpenBUGS and upload it to dscratch. First upload it to AFS and then copy it to dscratch. * Unzip the file on the cluster * Create a file called linbugs with the following content and change permission using chmod 755 lingbugs #!/bin/bash export LD_ASSUME_KERNEL=2.4.1 DIR=$(dirname $0) cd "$DIR" if [ \! -e "$DIR/temp" ] ; then mkdir "$DIR/temp" fi if [ -e bugs.so ] ; then ./cbugs "$DIR" "$DIR/temp" "/bugs.so" else ./cbugs "$DIR" "$DIR/temp" "/… text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/v1_on_nov_22_2008 /*********************************************************** *Mediation Analysis * * * * Author: Zhiyong Zhang * * * * Purpose: Test mediation through bootstrap and EM * * algorithm for missing data * * * * Usage: … text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/v1_on_nov_23_2008 /*********************************************************** *Mediation Analysis * * * * Author: Zhiyong Zhang * * * * Purpose: Test mediation through bootstrap and EM * * algorithm for missing data * * * * Usage: … text/html 2016-01-24T14:48:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://nd.psychstat.org/lab/zhang Papers undergoing * How vague of multivariate priors * PPP for model evaluation in Bayesian SEM * DIC for model comparison in Bayesian SEM * IRT with guessing * Dynamic mediation analysis with Sam * Robust mediation analysis * Power analysis with Timo & Manuel * Bayesian exploratory factor analysis using informative priors * Missing data mediation analysis * Bayesian differential equations by Johnny, Xin, Laura * Robust growth curve modeling using Student-t distribution by…