Loading required package: Rcpp Loading required package: inline Attaching package: ‘inline’ The following object is masked from ‘package:Rcpp’: registerPlugin rstan (Version 2.2.0, packaged: 2014-02-14 04:29:17 UTC, GitRev: 52d7b230aaa0) Loading required package: lattice Attaching package: ‘coda’ The following object is masked from ‘package:rstan’: traceplot Loading required package: boot Attaching package: ‘boot’ The following object is masked from ‘package:lattice’: melanoma Loading required package: MASS Loading required package: segmented mixtools package, version 1.0.1, Released January 2014 This package is based upon work supported by the National Science Foundation under Grant No. SES-0518772. **************** Cleaning data for K3 Simulation Stan unordered @ 3 Removing 0 of 10 Level 2 units for length. Calculating initial values for chain 1 ; K3 Simulation Stan unordered @ 3 number of iterations= 51 One of the variances is going to zero; trying new starting values. number of iterations= 110 number of iterations= 27 number of iterations= 73 number of iterations= 111 number of iterations= 293 number of iterations= 23 number of iterations= 47 number of iterations= 38 number of iterations= 82 Calculating initial values for chain 2 ; K3 Simulation Stan unordered @ 3 One of the variances is going to zero; trying new starting values. One of the variances is going to zero; trying new starting values. One of the variances is going to zero; trying new starting values. One of the variances is going to zero; trying new starting values. One of the variances is going to zero; trying new starting values. One of the variances is going to zero; trying new starting values. One of the variances is going to zero; trying new starting values. One of the variances is going to zero; trying new starting values. number of iterations= 101 number of iterations= 49 number of iterations= 14 number of iterations= 12 number of iterations= 31 number of iterations= 29 number of iterations= 21 number of iterations= 17 One of the variances is going to zero; trying new starting values. number of iterations= 70 number of iterations= 126 Calculating initial values for chain 3 ; K3 Simulation Stan unordered @ 3 number of iterations= 79 number of iterations= 249 number of iterations= 21 number of iterations= 13 number of iterations= 36 number of iterations= 48 number of iterations= 11 One of the variances is going to zero; trying new starting values. One of the variances is going to zero; trying new starting values. One of the variances is going to zero; trying new starting values. number of iterations= 39 number of iterations= 41 number of iterations= 219 **************** Running Model for K3 Simulation Stan unordered @ 3 Attempt 1 TRANSLATING MODEL 'hierModel1p' FROM Stan CODE TO C++ CODE NOW. COMPILING THE C++ CODE FOR MODEL 'hierModel1p' NOW. SAMPLING FOR MODEL 'hierModel1p' NOW (CHAIN 1). Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Iteration: 1 / 6000 [ 0%] (Warmup) Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Iteration: 600 / 6000 [ 10%] (Warmup) Iteration: 1200 / 6000 [ 20%] (Sampling) Iteration: 1800 / 6000 [ 30%] (Sampling) Iteration: 2400 / 6000 [ 40%] (Sampling) Iteration: 3000 / 6000 [ 50%] (Sampling) Iteration: 3600 / 6000 [ 60%] (Sampling) Iteration: 4200 / 6000 [ 70%] (Sampling) Iteration: 4800 / 6000 [ 80%] (Sampling) Iteration: 5400 / 6000 [ 90%] (Sampling) Iteration: 6000 / 6000 [100%] (Sampling) Elapsed Time: 64.805 seconds (Warm-up) 316.959 seconds (Sampling) 381.764 seconds (Total) SAMPLING FOR MODEL 'hierModel1p' NOW (CHAIN 2). Iteration: 1 / 6000 [ 0%] (Warmup) Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Iteration: 600 / 6000 [ 10%] (Warmup) Iteration: 1200 / 6000 [ 20%] (Sampling) Iteration: 1800 / 6000 [ 30%] (Sampling) Iteration: 2400 / 6000 [ 40%] (Sampling) Iteration: 3000 / 6000 [ 50%] (Sampling) Iteration: 3600 / 6000 [ 60%] (Sampling) Iteration: 4200 / 6000 [ 70%] (Sampling) Iteration: 4800 / 6000 [ 80%] (Sampling) Iteration: 5400 / 6000 [ 90%] (Sampling) Iteration: 6000 / 6000 [100%] (Sampling) Elapsed Time: 66.4396 seconds (Warm-up) 289.809 seconds (Sampling) 356.249 seconds (Total) SAMPLING FOR MODEL 'hierModel1p' NOW (CHAIN 3). Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Iteration: 1 / 6000 [ 0%] (Warmup) Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Location parameter is -inf:0, but must be finite! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Iteration: 600 / 6000 [ 10%] (Warmup) Iteration: 1200 / 6000 [ 20%] (Sampling) Iteration: 1800 / 6000 [ 30%] (Sampling) Iteration: 2400 / 6000 [ 40%] (Sampling) Iteration: 3000 / 6000 [ 50%] (Sampling) Iteration: 3600 / 6000 [ 60%] (Sampling) Iteration: 4200 / 6000 [ 70%] (Sampling) Iteration: 4800 / 6000 [ 80%] (Sampling) Iteration: 5400 / 6000 [ 90%] (Sampling) Iteration: 6000 / 6000 [100%] (Sampling) Elapsed Time: 98.8965 seconds (Warm-up) 491.589 seconds (Sampling) 590.486 seconds (Total) Labeling components for level 2 model K3 Simulation Stan unordered @ 3 Labeling components for alpha0 Labeling components for mu0 Labeling components for beta0 Labeling components for tau0 Labeling components for gamma0 Labeling components for pi Labeling components for mu Labeling components for sigma **************** Convergence diagnostics for K3 Simulation Stan unordered @ 3 Run Number 1 Iterations = 1:5000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 5000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE -616.76246 10.47600 0.08554 0.24924 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% -638.1 -623.7 -616.5 -609.6 -597.0 Potential scale reduction factors: Point est. Upper C.I. lp__ 1.04 1.14 lp__ 1708.282 Iterations = 1:5000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 5000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE 13.30699 4.58207 0.03741 0.05787 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% 6.04 10.02 12.75 16.00 23.80 Potential scale reduction factors: Point est. Upper C.I. alphaN 1.03 1.1 alphaN 6904.41 Iterations = 1:5000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 5000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE alpha0[1] 0.4131 0.06807 0.0005558 0.0010255 alpha0[2] 0.4128 0.06296 0.0005141 0.0008883 alpha0[3] 0.1740 0.05975 0.0004879 0.0012840 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% alpha0[1] 0.27494 0.3676 0.4147 0.4584 0.5439 alpha0[2] 0.28929 0.3717 0.4123 0.4543 0.5363 alpha0[3] 0.08133 0.1307 0.1655 0.2084 0.3114 Potential scale reduction factors: Point est. Upper C.I. alpha0[2] 1.00 1.00 alpha0[3] 1.14 1.41 Multivariate psrf 1.14 alpha0[1] alpha0[2] alpha0[3] 4729.853 5421.687 2114.486 Iterations = 1:5000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 5000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE mu0[1] -0.8429 0.2159 0.001763 0.004618 mu0[2] -0.3842 0.2157 0.001761 0.003485 mu0[3] 1.3577 0.4991 0.004075 0.009991 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% mu0[1] -1.2044 -1.0192 -0.8433 -0.6834 -0.43331 mu0[2] -0.7716 -0.5296 -0.3954 -0.2516 0.07141 mu0[3] 0.3115 1.0281 1.3942 1.7160 2.20931 Potential scale reduction factors: Point est. Upper C.I. mu0[1] 1.53 2.36 mu0[2] 1.06 1.19 mu0[3] 1.06 1.19 Multivariate psrf 1.5 mu0[1] mu0[2] mu0[3] 2370.169 3800.723 2391.535 Iterations = 1:5000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 5000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE beta0[1] 0.4807 0.2454 0.002004 0.007443 beta0[2] 0.6345 0.1975 0.001613 0.003411 beta0[3] 0.6890 0.3873 0.003162 0.008481 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% beta0[1] 0.09231 0.2988 0.4736 0.6329 1.007 beta0[2] 0.35094 0.5012 0.6015 0.7278 1.116 beta0[3] 0.13739 0.3975 0.6249 0.9111 1.578 Potential scale reduction factors: Point est. Upper C.I. beta0[1] 1.67 2.64 beta0[2] 1.01 1.01 beta0[3] 1.00 1.00 Multivariate psrf 1.5 beta0[1] beta0[2] beta0[3] 1262.441 4982.193 2393.099 Iterations = 1:5000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 5000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE tau0[1] 1.538 0.8050 0.006573 0.01813 tau0[2] 2.077 0.9506 0.007762 0.03358 tau0[3] -0.269 0.6127 0.005003 0.01323 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% tau0[1] 0.04465 0.9917 1.4993 2.03489 3.281 tau0[2] 0.20194 1.3905 2.1236 2.82378 3.697 tau0[3] -1.23803 -0.7040 -0.3562 0.08377 1.124 Potential scale reduction factors: Point est. Upper C.I. tau0[1] 1.04 1.14 tau0[2] 1.13 1.40 tau0[3] 1.06 1.19 Multivariate psrf 1.15 tau0[1] tau0[2] tau0[3] 2543.7611 726.1588 2594.2593 Iterations = 1:5000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 5000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE gamma0[1] 2.146 0.6227 0.005084 0.01100 gamma0[2] 1.907 0.6862 0.005603 0.01903 gamma0[3] 1.319 0.6255 0.005107 0.01277 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% gamma0[1] 1.1709 1.7214 2.056 2.474 3.597 gamma0[2] 0.8329 1.4271 1.821 2.289 3.549 gamma0[3] 0.4506 0.8487 1.203 1.675 2.818 Potential scale reduction factors: Point est. Upper C.I. gamma0[1] 1.01 1.02 gamma0[2] 1.06 1.20 gamma0[3] 1.09 1.28 Multivariate psrf 1.13 gamma0[1] gamma0[2] gamma0[3] 3720.001 1391.854 2253.634 Chains of length 5000 for K3 Simulation Stan unordered @ 3 did not converge in run 1 . Maximum Rhat value = 1.500166 . lp__ [[ 1 ]] Mean SD Naive SE Time-series SE -619.6171151 10.4124597 0.1472544 0.4412830 lp__ [[ 2 ]] Mean SD Naive SE Time-series SE -615.7258074 10.3320306 0.1461170 0.4111837 lp__ [[ 3 ]] Mean SD Naive SE Time-series SE -614.9444479 10.0787957 0.1425357 0.4419317 alphaN [[ 1 ]] Mean SD Naive SE Time-series SE 12.24322950 4.33935478 0.06136774 0.12201440 alphaN [[ 2 ]] Mean SD Naive SE Time-series SE 13.79608517 4.55220496 0.06437790 0.08162934 alphaN [[ 3 ]] Mean SD Naive SE Time-series SE 13.88164907 4.66266580 0.06594005 0.09270504 alpha0 [[ 1 ]] Mean SD Naive SE Time-series SE alpha0[1] 0.3864068 0.06756858 0.0009555640 0.002273420 alpha0[2] 0.4124028 0.06005367 0.0008492871 0.001257084 alpha0[3] 0.2011904 0.06380767 0.0009023767 0.002723708 alpha0 [[ 2 ]] Mean SD Naive SE Time-series SE alpha0[1] 0.4270223 0.06536913 0.0009244592 0.001473331 alpha0[2] 0.4133796 0.06661240 0.0009420417 0.001854598 alpha0[3] 0.1595980 0.05438198 0.0007690773 0.002155746 alpha0 [[ 3 ]] Mean SD Naive SE Time-series SE alpha0[1] 0.4259131 0.06318920 0.0008936303 0.001458140 alpha0[2] 0.4127637 0.06205467 0.0008775855 0.001442960 alpha0[3] 0.1613232 0.05075000 0.0007177133 0.001664789 mu0 [[ 1 ]] Mean SD Naive SE Time-series SE mu0[1] -1.0158385 0.1503759 0.002126636 0.011198902 mu0[2] -0.3166227 0.2120658 0.002999063 0.005979147 mu0[3] 1.1978327 0.4804137 0.006794075 0.017287893 mu0 [[ 2 ]] Mean SD Naive SE Time-series SE mu0[1] -0.7474484 0.1900344 0.002687493 0.005485041 mu0[2] -0.4272561 0.2037789 0.002881869 0.006727856 mu0[3] 1.4492850 0.4996708 0.007066412 0.018536323 mu0 [[ 3 ]] Mean SD Naive SE Time-series SE mu0[1] -0.7654309 0.1898931 0.002685494 0.006036651 mu0[2] -0.4086797 0.2145530 0.003034237 0.005321812 mu0[3] 1.4259642 0.4779536 0.006759285 0.015997130 beta0 [[ 1 ]] Mean SD Naive SE Time-series SE beta0[1] 0.2675990 0.1756948 0.002484700 0.018956721 beta0[2] 0.6224272 0.1832442 0.002591464 0.004002747 beta0[3] 0.6773111 0.3664392 0.005182233 0.011418430 beta0 [[ 2 ]] Mean SD Naive SE Time-series SE beta0[1] 0.5994198 0.2090297 0.002956126 0.008127535 beta0[2] 0.6379930 0.2140780 0.003027520 0.008490911 beta0[3] 0.6900275 0.3974373 0.005620612 0.013229169 beta0 [[ 3 ]] Mean SD Naive SE Time-series SE beta0[1] 0.5751799 0.1941264 0.002745361 0.008553485 beta0[2] 0.6429961 0.1934423 0.002735687 0.004077446 beta0[3] 0.6996432 0.3968629 0.005612489 0.018490623 tau0 [[ 1 ]] Mean SD Naive SE Time-series SE tau0[1] 1.3316453 0.6597754 0.009330634 0.01773809 tau0[2] 2.5119426 0.7927424 0.011211070 0.04889120 tau0[3] -0.4618085 0.5110294 0.007227047 0.01512779 tau0 [[ 2 ]] Mean SD Naive SE Time-series SE tau0[1] 1.6728758 0.8450972 0.01195148 0.03152930 tau0[2] 1.8265446 0.9463612 0.01338357 0.06236882 tau0[3] -0.1778415 0.6242233 0.00882785 0.01956274 tau0 [[ 3 ]] Mean SD Naive SE Time-series SE tau0[1] 1.6107933 0.8538130 0.012074739 0.04061946 tau0[2] 1.8927543 0.9493713 0.013426138 0.06219372 tau0[3] -0.1673247 0.6478354 0.009161776 0.03106333 gamma0 [[ 1 ]] Mean SD Naive SE Time-series SE gamma0[1] 2.214494 0.5929675 0.008385827 0.01408161 gamma0[2] 1.683325 0.6303388 0.008914336 0.03105437 gamma0[3] 1.083274 0.4967780 0.007025503 0.01911896 gamma0 [[ 2 ]] Mean SD Naive SE Time-series SE gamma0[1] 2.104871 0.6455945 0.009130085 0.02394333 gamma0[2] 2.020043 0.6763651 0.009565247 0.03999133 gamma0[3] 1.435609 0.6380674 0.009023636 0.02108213 gamma0 [[ 3 ]] Mean SD Naive SE Time-series SE gamma0[1] 2.118493 0.6228407 0.008808298 0.01783351 gamma0[2] 2.018809 0.6947689 0.009825516 0.02637675 gamma0[3] 1.439011 0.6606066 0.009342389 0.02566606 **************** Running Model for K3 Simulation Stan unordered @ 3 Attempt 2 SAMPLING FOR MODEL 'hierModel1p' NOW (CHAIN 1). Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Iteration: 1 / 12000 [ 0%] (Warmup) Iteration: 1200 / 12000 [ 10%] (Warmup) Iteration: 2400 / 12000 [ 20%] (Sampling) Iteration: 3600 / 12000 [ 30%] (Sampling) Iteration: 4800 / 12000 [ 40%] (Sampling) Iteration: 6000 / 12000 [ 50%] (Sampling) Iteration: 7200 / 12000 [ 60%] (Sampling) Iteration: 8400 / 12000 [ 70%] (Sampling) Iteration: 9600 / 12000 [ 80%] (Sampling) Iteration: 10800 / 12000 [ 90%] (Sampling) Iteration: 12000 / 12000 [100%] (Sampling) Elapsed Time: 125.786 seconds (Warm-up) 693.024 seconds (Sampling) 818.81 seconds (Total) SAMPLING FOR MODEL 'hierModel1p' NOW (CHAIN 2). Iteration: 1 / 12000 [ 0%] (Warmup) Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Iteration: 1200 / 12000 [ 10%] (Warmup) Iteration: 2400 / 12000 [ 20%] (Sampling) Iteration: 3600 / 12000 [ 30%] (Sampling) Iteration: 4800 / 12000 [ 40%] (Sampling) Iteration: 6000 / 12000 [ 50%] (Sampling) Iteration: 7200 / 12000 [ 60%] (Sampling) Iteration: 8400 / 12000 [ 70%] (Sampling) Iteration: 9600 / 12000 [ 80%] (Sampling) Iteration: 10800 / 12000 [ 90%] (Sampling) Iteration: 12000 / 12000 [100%] (Sampling) Elapsed Time: 106.315 seconds (Warm-up) 491.549 seconds (Sampling) 597.864 seconds (Total) SAMPLING FOR MODEL 'hierModel1p' NOW (CHAIN 3). Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Iteration: 1 / 12000 [ 0%] (Warmup) Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Location parameter is -inf:0, but must be finite! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Informational Message: The current Metropolis proposal is about to be rejected becuase of the following issue: Error in function stan::prob::normal_log(N4stan5agrad3varE): Scale parameter is 0:0, but must be > 0! If this warning occurs sporadically, such as for highly constrained variable types like covariance matrices, then the sampler is fine, but if this warning occurs often then your model may be either severely ill-conditioned or misspecified. Iteration: 1200 / 12000 [ 10%] (Warmup) Iteration: 2400 / 12000 [ 20%] (Sampling) Iteration: 3600 / 12000 [ 30%] (Sampling) Iteration: 4800 / 12000 [ 40%] (Sampling) Iteration: 6000 / 12000 [ 50%] (Sampling) Iteration: 7200 / 12000 [ 60%] (Sampling) Iteration: 8400 / 12000 [ 70%] (Sampling) Iteration: 9600 / 12000 [ 80%] (Sampling) Iteration: 10800 / 12000 [ 90%] (Sampling) Iteration: 12000 / 12000 [100%] (Sampling) Elapsed Time: 125.246 seconds (Warm-up) 394.371 seconds (Sampling) 519.617 seconds (Total) Labeling components for level 2 model K3 Simulation Stan unordered @ 3 Labeling components for alpha0 Labeling components for mu0 Labeling components for beta0 Labeling components for tau0 Labeling components for gamma0 Labeling components for pi Labeling components for mu Labeling components for sigma **************** Convergence diagnostics for K3 Simulation Stan unordered @ 3 Run Number 2 Iterations = 1:10000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 10000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE -614.88360 10.11160 0.05838 0.16175 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% -635.7 -621.5 -614.6 -607.8 -596.2 Potential scale reduction factors: Point est. Upper C.I. lp__ 1 1 lp__ 4049.295 Iterations = 1:10000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 10000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE 13.57127 4.63606 0.02677 0.04128 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% 6.356 10.185 12.946 16.313 24.280 Potential scale reduction factors: Point est. Upper C.I. alphaN 1.03 1.1 alphaN 13032.16 Iterations = 1:15000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 15000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE alpha0[1] 0.4266 0.06367 0.0003002 0.0009676 alpha0[2] 0.3925 0.07516 0.0003543 0.0022299 alpha0[3] 0.1809 0.06923 0.0003264 0.0013408 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% alpha0[1] 0.29784 0.3851 0.4277 0.4696 0.5493 alpha0[2] 0.22478 0.3478 0.3979 0.4433 0.5273 alpha0[3] 0.08045 0.1305 0.1685 0.2190 0.3484 Potential scale reduction factors: Point est. Upper C.I. alpha0[2] 1.13 1.38 alpha0[3] 1.31 1.83 Multivariate psrf 1.25 alpha0[1] alpha0[2] alpha0[3] 7737.125 5525.038 2982.202 Iterations = 1:15000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 15000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE mu0[1] -0.9234 0.1692 0.0007977 0.008037 mu0[2] -0.1834 0.2849 0.0013431 0.012760 mu0[3] 1.2396 0.5645 0.0026610 0.013238 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% mu0[1] -1.1907 -1.0352 -0.9512 -0.83744 -0.5177 mu0[2] -0.7154 -0.3845 -0.1872 0.00559 0.3908 mu0[3] 0.1625 0.8171 1.2778 1.67207 2.2052 Potential scale reduction factors: Point est. Upper C.I. mu0[1] 1.04 1.14 mu0[2] 1.04 1.14 mu0[3] 1.24 1.67 Multivariate psrf 1.23 mu0[1] mu0[2] mu0[3] 567.6133 493.9351 2508.4469 Iterations = 1:15000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 15000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE beta0[1] 0.3640 0.1949 0.0009186 0.013598 beta0[2] 0.6528 0.2712 0.0012787 0.004631 beta0[3] 0.7723 0.4225 0.0019919 0.009044 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% beta0[1] 0.1312 0.2335 0.3066 0.4434 0.8676 beta0[2] 0.2890 0.4775 0.5995 0.7601 1.3472 beta0[3] 0.1581 0.4465 0.7127 1.0291 1.7346 Potential scale reduction factors: Point est. Upper C.I. beta0[1] 1.06 1.20 beta0[2] 1.09 1.22 beta0[3] 1.09 1.27 Multivariate psrf 1.16 beta0[1] beta0[2] beta0[3] 462.7261 3480.7477 4057.7878 Iterations = 1:15000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 15000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE tau0[1] 1.90246 0.7919 0.003733 0.02339 tau0[2] 1.43592 0.9584 0.004518 0.07122 tau0[3] 0.01507 0.8225 0.003877 0.03972 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% tau0[1] 0.2671 1.3763 1.9443 2.4671 3.328 tau0[2] -0.3916 0.7961 1.3846 2.0377 3.410 tau0[3] -1.2372 -0.5838 -0.1109 0.4884 1.980 Potential scale reduction factors: Point est. Upper C.I. tau0[1] 1.00 1.01 tau0[2] 1.03 1.06 tau0[3] 1.15 1.43 Multivariate psrf 1.11 tau0[1] tau0[2] tau0[3] 1665.367 673.160 2213.323 Iterations = 1:15000 Thinning interval = 1 Number of chains = 3 Sample size per chain = 15000 1. Empirical mean and standard deviation for each variable, plus standard error of the mean: Mean SD Naive SE Time-series SE gamma0[1] 2.144 0.6432 0.003032 0.006814 gamma0[2] 2.029 0.6539 0.003082 0.014159 gamma0[3] 1.566 0.7173 0.003382 0.031854 2. Quantiles for each variable: 2.5% 25% 50% 75% 97.5% gamma0[1] 1.1888 1.701 2.042 2.471 3.682 gamma0[2] 0.9363 1.596 1.950 2.380 3.563 gamma0[3] 0.4885 1.021 1.491 1.993 3.207 Potential scale reduction factors: Point est. Upper C.I. gamma0[1] 1.00 1.00 gamma0[2] 1.00 1.01 gamma0[3] 1.04 1.13 Multivariate psrf 1.04 gamma0[1] gamma0[2] gamma0[3] 9313.606 3476.199 3208.698 Chains of length 10000 for K3 Simulation Stan unordered @ 3 did not converge in run 2 . Maximum Rhat value = 1.249106 . lp__ [[ 1 ]] Mean SD Naive SE Time-series SE -614.54009201 9.76342202 0.09763422 0.24091651 lp__ [[ 2 ]] Mean SD Naive SE Time-series SE -615.1568706 10.1976433 0.1019764 0.2807325 lp__ [[ 3 ]] Mean SD Naive SE Time-series SE -614.9538455 10.3556778 0.1035568 0.3140377 alphaN [[ 1 ]] Mean SD Naive SE Time-series SE 12.48848291 4.22392296 0.04223923 0.05777725 alphaN [[ 2 ]] Mean SD Naive SE Time-series SE 14.05403403 4.71595976 0.04715960 0.07065654 alphaN [[ 3 ]] Mean SD Naive SE Time-series SE 14.17129461 4.75771811 0.04757718 0.08367963 alpha0 [[ 1 ]] Mean SD Naive SE Time-series SE alpha0[1] 0.4157984 0.06835481 0.0005581147 0.002516967 alpha0[2] 0.3598621 0.08189726 0.0006686883 0.006437694 alpha0[3] 0.2243394 0.07276210 0.0005941001 0.003321494 alpha0 [[ 2 ]] Mean SD Naive SE Time-series SE alpha0[1] 0.4324096 0.06055383 0.0004944199 0.001013841 alpha0[2] 0.4075017 0.06644899 0.0005425537 0.001392763 alpha0[3] 0.1600887 0.05659043 0.0004620589 0.001747279 alpha0 [[ 3 ]] Mean SD Naive SE Time-series SE alpha0[1] 0.4316260 0.06040175 0.0004931782 0.001031002 alpha0[2] 0.4102505 0.06496288 0.0005304197 0.001169233 alpha0[3] 0.1581235 0.05517291 0.0004504849 0.001447244 mu0 [[ 1 ]] Mean SD Naive SE Time-series SE mu0[1] -0.9684290 0.1435294 0.001171913 0.007953424 mu0[2] -0.1093595 0.3105073 0.002535282 0.025598871 mu0[3] 0.9043139 0.5051192 0.004124281 0.033280203 mu0 [[ 2 ]] Mean SD Naive SE Time-series SE mu0[1] -0.9021052 0.1776050 0.001450138 0.01678516 mu0[2] -0.2190813 0.2624596 0.002142973 0.02111596 mu0[3] 1.4062046 0.5191605 0.004238928 0.01590489 mu0 [[ 3 ]] Mean SD Naive SE Time-series SE mu0[1] -0.8997536 0.1752687 0.001431063 0.01537282 mu0[2] -0.2218870 0.2646149 0.002160572 0.01908135 mu0[3] 1.4084147 0.5124428 0.004184078 0.01471583 beta0 [[ 1 ]] Mean SD Naive SE Time-series SE beta0[1] 0.3015564 0.1595321 0.001302574 0.008465941 beta0[2] 0.7238149 0.3454109 0.002820268 0.010763003 beta0[3] 0.9291315 0.4459400 0.003641085 0.023961968 beta0 [[ 2 ]] Mean SD Naive SE Time-series SE beta0[1] 0.3921809 0.2080824 0.001698986 0.031102548 beta0[2] 0.6101687 0.2158724 0.001762591 0.006484469 beta0[3] 0.6906967 0.3875210 0.003164096 0.009230504 beta0 [[ 3 ]] Mean SD Naive SE Time-series SE beta0[1] 0.3982300 0.1982421 0.001618640 0.025001650 beta0[2] 0.6244083 0.2171456 0.001772987 0.005926321 beta0[3] 0.6971790 0.3869531 0.003159459 0.008765234 tau0 [[ 1 ]] Mean SD Naive SE Time-series SE tau0[1] 1.838072 0.8019741 0.006548091 0.05655172 tau0[2] 1.318853 1.1567633 0.009444932 0.20319272 tau0[3] 0.369909 0.9675045 0.007899642 0.11553203 tau0 [[ 2 ]] Mean SD Naive SE Time-series SE tau0[1] 1.9337136 0.7724204 0.006306786 0.02674674 tau0[2] 1.4861486 0.8335710 0.006806079 0.04653814 tau0[3] -0.1695221 0.6740956 0.005503968 0.02133918 tau0 [[ 3 ]] Mean SD Naive SE Time-series SE tau0[1] 1.9355882 0.7972259 0.006509322 0.03175569 tau0[2] 1.5027457 0.8378214 0.006840783 0.04684103 tau0[3] -0.1551863 0.6710466 0.005479072 0.01982871 gamma0 [[ 1 ]] Mean SD Naive SE Time-series SE gamma0[1] 2.147850 0.6430383 0.005250386 0.01108869 gamma0[2] 1.981751 0.6797606 0.005550222 0.03574109 gamma0[3] 1.748650 0.7816572 0.006382204 0.09244888 gamma0 [[ 2 ]] Mean SD Naive SE Time-series SE gamma0[1] 2.147847 0.6483881 0.005294067 0.01064753 gamma0[2] 2.054912 0.6374492 0.005204751 0.01570814 gamma0[3] 1.468816 0.6566078 0.005361180 0.01522855 gamma0 [[ 3 ]] Mean SD Naive SE Time-series SE gamma0[1] 2.136920 0.6379685 0.005208991 0.01347222 gamma0[2] 2.049150 0.6410611 0.005234242 0.01673343 gamma0[3] 1.481891 0.6721241 0.005487870 0.01880008 MCMC run did not converge, proceeding anyway. Calculating model fit indexes for K3 Simulation Stan unordered @ 3 lppd pWAIC1 WAIC1 pWAIC2 WAIC2 -533.29001 60.02722 1186.63445 60.02722 1186.63445 lppd lppd.bayes pDIC DIC pDICalt DICalt -563.3036 -683.5002 -240.3932 886.2141 117.2020 1601.4045 Analaysis complete for K3 Simulation Stan unordered @ 3 > proc.time() user system elapsed 3473.781 9.698 3485.561