《140656 Multi-Level Statistical Models》由会员分享,可在线阅读,更多相关《140656 Multi-Level Statistical Models(56页珍藏版)》请在金锄头文库上搜索。
1、Term 4, 2006,BIO656-Multilevel Models,1,140.656 Multi-Level Statistical Models,If you did not receive the welcome email from me, email me at: (tlouisjhsph.edu),Term 4, 2006,BIO656-Multilevel Models,2,ROOM CHANGE, AGAIN!,Starting Thursday, March 30th and henceforth, lectures will be in W2030Labs will
2、 still be in W2009,Term 4, 2006,BIO656-Multilevel Models,3,Term 4, 2006,BIO656-Multilevel Models,4,Prerequisites, resources and Grading,Term 4, 2006,BIO656-Multilevel Models,5,Learning Objectives,Term 4, 2006,BIO656-Multilevel Models,6,Content & Approach,Term 4, 2006,BIO656-Multilevel Models,7,Appro
3、ach,Lectures include basic illustrations and case studies, structuring an approach and interpreting resultsLabs address computing and amplify on the foregoingMy approach is formal, but not “mathematical”To understand MLMs, you need a very good understanding on single-level modelsIf you understand th
4、ese, you are ready to multi-level!,Term 4, 2006,BIO656-Multilevel Models,8,Structure,Term 4, 2006,BIO656-Multilevel Models,9,RULES FOR HOMEWORK,MID-TERM AND PROJECT,Homework Must be individually prepared, but you can get help Homework due dates should be honored. Turn in hard copy for grading The in
5、-class, midterm Must be prepared absolutely independently During the exam, no advice or information can be obtained from others You can use your notes and reference materials The term project Must be individually prepared, but you can get help Must be electronically submitted,Term 4, 2006,BIO656-Mul
6、tilevel Models,10,Handouts and the Web,Virtually all course materials will be on the web Check frequently for updates Ive provided hard copy of the general information sheet However, other lectures will be on the web in powerpoint format and wont be handed out Download to your computer so you have a
7、n electronic version each part Print if you need hard copy, but do it 4 or 6 to a page to save paper More generally, try to “go electronic” printing sparingly,Term 4, 2006,BIO656-Multilevel Models,11,COMPUTING & DATA,We will support WinBUGS, StataWe provide partial support for SAS, which should be u
8、sed only by current SAS users; we arent teaching it from scratchSome homeworks require use of WinBUGS and another “traditional” program (STATA, SAS, R,.)We provide datasets, including some in the WinBUGS examples,Term 4, 2006,BIO656-Multilevel Models,12,WHY BUGS?,Freeware!In MLMs, its important to s
9、ee distributions e.g., Skewness of sampling distribution of variance component estimatesIts important to incorporate all uncertainties in estimating random effectsNote that WinBugs isnt very data input friendlyAnd, its difficult to produce P-values ,Term 4, 2006,BIO656-Multilevel Models,13,STATISTIC
10、AL MODELS,A statistical model is an approximation Almost never is there a “correct” or “best” model, no holy grailA model is a tool for structuring a statistical approach and addressing a scientific questionAn effective model combines the data with prior information to address a question,Term 4, 200
11、6,BIO656-Multilevel Models,14,MULTI-LEVEL MODELS,Biological, physical, psycho/social processes that influence health occur at many levels:Cell Organ Person Family Nhbd City Society . Solar systemCrew VesselFleet .Block Block Group Tract .Visit Patient Phy Clinic HMO .Covariates can be at each levelM
12、any “units of analysis”More modern and flexible parlance and approach: “many variance components”,Term 4, 2006,BIO656-Multilevel Models,15,Example: Alcohol Abuse,Cell: neurochemistry Organ: ability to metabolize ethanolPerson: genetic susceptibility to addictionFamily: alcohol abuse in the homeNeigh
13、borhood: availability of barsSociety: regulations; organizations; social norms,Term 4, 2006,BIO656-Multilevel Models,16,ALCOHOL ABUSE:A multi-level, interaction model,Interaction between existence of bars & state, drunk driving lawsAlcohol abuse in a family & ability to metabolize ethanolGenetic pre
14、disposition to addiction & household environmentState regulations about intoxication & job requirements,Term 4, 2006,BIO656-Multilevel Models,17,Many names for similar, but not identical models, analyses and goals,Multi-Level ModelsRandom effects modelsMixed modelsRandom coefficient modelsHierarchic
15、al modelsBayesian Models,Term 4, 2006,BIO656-Multilevel Models,18,We dont need MLMs,If your question is about slopes on regressors, you can run a standard regression and (usually) get valid slope estimatesY = 0 + 1(areal monitor) + 2(home monitor) + .Y = 0 + 1(zipcode income) + 2(personal income) +
16、.logit(P) = .Analysis can be followed by computing a “robust” SE to get valid inferences,Term 4, 2006,BIO656-Multilevel Models,19,We do need MLMs,If your question is about variance components, you need to build the multi-level modelYijkl = 0 + 1X1 + 2 X2 + . + ijkl Var(Yijkl) = Var(ijkl) = = VHospital + VClinic + VPhysician + VPatient + VunexplainedThese variances depend on what Xs are in the model,