| Chapter 52 TheMCMCProcedure7 p: f8 s: P; o# ~ Contents$ ?6 `+ U8 u. d$ L5 [1 X; q6 Q Overview:MCMCProcedure ............................ 3478) r @, |% C2 L PROCMCMCComparedwithOtherSASProcedures ............3479 GettingStarted:MCMCProcedure .......................... 3479 SimpleLinearRegression ...........................3480* t8 F3 F G( v TheBehrens-FisherProblem ..........................34888 V5 N; ?* r: e" \2 v6 N Mixed-EffectsModel .............................3492! @% X O0 I2 g Syntax:MCMCProcedure .............................. 3495 PROCMCMCStatement ...........................3496 ARRAYStatement ...............................3508" @7 k* ~& `* B/ L BEGINCNST/ENDCNSTStatement .....................3509' O+ V- ?6 A- R BEGINNODATA/ENDNODATAStatements .................3511 BYStatement .................................3511' t+ R$ r) {8 i7 h# { MODELStatement ...............................3512/ M2 F4 c& Z) n PARMSStatement ...............................3515# d. ^& L5 U; N; A# d2 Q/ m PRIOR/HYPERPRIORStatement .......................3516 ProgrammingStatements ...........................3516 UDSStatement .................................3518 Details:MCMCProcedure .............................. 3522 HowPROCMCMCWorks ..........................3522 BlockingofParameters ............................3523 Samplers ....................................35242 S+ |5 a# \* J0 V% U TuningtheProposalDistribution .......................3525: m+ e7 W A6 C( h* @' u# H InitialValuesoftheMarkovChains ......................35288 m( o" w8 c- W4 {. J! w% u1 a2 N AssignmentsofParameters ..........................3528 g: K% |+ Q$ ?% J( G( k StandardDistributions .............................35305 p9 r9 `; O7 k7 @/ R, a3 J# z SpecifyingaNewDistribution .........................3541 UsingDensityFunctionsintheProgrammingStatements ...........3542( }* y) n5 M8 r$ j0 h TruncationandCensoring ...........................35446 J h1 Z. [# Y, | G& h& f MultivariateDensityFunctions ........................3546# g/ {8 o2 M% ]9 H# f: {6 o SomeUsefulSASFunctions ..........................3549 MatrixFunctionsinPROCMCMC ......................3551, G- j* u) l: v ModelingJointLikelihood ...........................3556- S6 h$ z$ e6 `7 n$ e RegeneratingDiagnosticsPlots ........................3557- Y/ w2 T) x/ f5 _2 P PosteriorPredictiveDistribution ........................35604 s7 {. {+ l0 ~' B* W HandlingofMissingData ...........................3565 FloatingPointErrorsandOverflows ......................3565 HandlingErrorMessages ...........................3568 ComputationalResources ...........................3570 DisplayedOutput ................................3571 ODSTableNames ...............................3575 ODSGraphics .................................3577" a' t' r* I: |. _: F0 v5 ~ Examples:MCMCProcedure ............................ 3578# N+ m4 X8 M8 i Example52.1:SimulatingSamplesFromaKnownDensity .........3578 Example52.2:Box-CoxTransformation ...................35836 I y3 W* H$ F+ z' \ Example52.3:GeneralizedLinearModels ..................3592/ A: B8 n6 c6 ]- V) p3 x T; s Example52.4:NonlinearPoissonRegressionModels ............3605. |5 y7 h5 u0 S0 ~8 ` Example52.5:Random-EffectsModels ...................3614 Example52.6:ChangePointModels .....................3630 Example52.7:ExponentialandWeibullSurvivalAnalysis ..........36348 n$ B e% C* L5 q' J Example52.8:CoxModels ..........................3647: k3 w1 m" f) o# Q. w Example52.9:NormalRegressionwithIntervalCensoring .........3664 Example52.10:ConstrainedAnalysis ....................36667 l/ n7 ~ h: \, Q+ e1 i! m: r0 @ Example52.11:ImplementaNewSamplingAlgorithm ...........36720 P. [2 ]5 \. _4 I( P Example52.12:UsingaTransformationtoImproveMixing .........3683 Example52.13:Gelman-RubinDiagnostics .................3693: W( e9 s* @1 p3 e* U1 ? References ...................................... 3700 |
| 欢迎光临 数学建模社区-数学中国 (http://www.madio.net/) | Powered by Discuz! X2.5 |