Last data update: 2014.03.03

 R: EM-like algorithm for the GMCM
PseudoEMAlgorithmR Documentation

EM-like algorithm for the GMCM

Description

An fast and modified implementation of the Li et. al. (2011) EM-like algorithm for estimating the maximizing parameters of the GMCM-likelihood function.

Usage

PseudoEMAlgorithm(x, theta, eps = 1e-04, max.ite = 1000, verbose = FALSE,
  trace.theta = FALSE, meta.special.case = FALSE,
  convergence.criterion = c("absGMCM", "GMCM", "GMM", "Li", "absLi"))

Arguments

x

A matrix of observations where rows corresponds to features and columns to experiments.

theta

A list of parameters formatted as described in rtheta.

eps

The maximum difference required to achieve convergence.

max.ite

The maximum number of iterations.

verbose

Logical. Set to TRUE to increase verbosity.

trace.theta

Logical. If TRUE, a trace of the estimated thetas are returned.

meta.special.case

Logical. If TRUE, the estimators used are for the special case GMCM of Li et. al. (2011).

convergence.criterion

Character. Sets the convergence criterion. If "absGMCM" the absolute value of difference in GMCM is used. If "GMCM" the difference in GMCM-likelihoods are used as convergence criterion. If "GMM", the guaranteed non-decreasing difference of GMM-likelihoods are used. If "Li", the convergence criterion used by Li et. al. (2011) is used. If "absLi", the absolute values of the Li et. al. criterion.

Details

When either "absGMCM" or "absLi" are used, the parameters corresponding to the biggest observed likelihood is returned. This is not necessarily the last iteration.

Value

A list of 3 or 4 is returned depending on the value of trace.theta

theta

A list containing the final parameter estimate in the format of rtheta

loglik.tr

A matrix with different log-likelihood traces in each row.

kappa

A matrix where the (i,j)'th entry is the probability that x[i, ] belongs to the j'th component. Usually the returned value of EStep.

theta.tr

A list of each obtained parameter estimates in the format of rtheta

Note

The algorithm is highly sensitive to the starting parameters which therefore should be carefully chosen.

Author(s)

Anders Ellern Bilgrau <anders.ellern.bilgrau@gmail.com>

References

Li, Q., Brown, J. B. J. B., Huang, H., & Bickel, P. J. (2011). Measuring reproducibility of high-throughput experiments. The Annals of Applied Statistics, 5(3), 1752-1779. doi:10.1214/11-AOAS466

See Also

rtheta, EMAlgorithm

Examples

set.seed(1)

# Choosing the true parameters and simulating data
true.par <- c(0.8, 3, 1.5, 0.4)
data <- SimulateGMCMData(n = 1000, par = true.par, d = 2)
uhat <- Uhat(data$u)  # Observed ranks

# Plot of latent and observed data colour coded by the true component
par(mfrow = c(1,2))
plot(data$z, main = "Latent data", cex = 0.6,
     xlab = "z (Experiment 1)", ylab = "z (Experiment 2)",
     col = c("red","blue")[data$K])
plot(uhat, main = "Observed data", cex = 0.6,
     xlab = "u (Experiment 1)", ylab = "u (Experiment 2)",
     col = c("red","blue")[data$K])

# Fit the model using the Pseudo EM algorithm
init.par <- c(0.5, 1, 1, 0.5)
res <- GMCM:::PseudoEMAlgorithm(uhat, meta2full(init.par, d = 2),
                                verbose = TRUE,
                                convergence.criterion = "absGMCM",
                                eps = 1e-4,
                                trace.theta = FALSE,
                                meta.special.case = TRUE)

# Compute posterior cluster probabilities
IDRs <- get.IDR(uhat, par = full2meta(res$theta))

# Plot of observed data colour coded by the MAP estimate
plot(res$loglik[3,], main = "Loglikelihood trace", type = "l",
     ylab = "log GMCM likelihood")
abline(v = which.max(res$loglik[3,])) # Chosen MLE
plot(uhat, main = "Clustering\nIDR < 0.05", xlab = "", ylab = "", cex = 0.6,
     col = c("Red","Blue")[IDRs$Khat])

# View parameters
rbind(init.par, true.par, estimate = full2meta(res$theta))

# Confusion matrix
table("Khat" = IDRs$Khat, "K" = data$K)

Results


R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(GMCM)
> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/GMCM/PseudoEMAlgorithm.Rd_%03d_medium.png", width=480, height=480)
> ### Name: PseudoEMAlgorithm
> ### Title: EM-like algorithm for the GMCM
> ### Aliases: PseudoEMAlgorithm
> ### Keywords: internal
> 
> ### ** Examples
> 
> set.seed(1)
> 
> # Choosing the true parameters and simulating data
> true.par <- c(0.8, 3, 1.5, 0.4)
> data <- SimulateGMCMData(n = 1000, par = true.par, d = 2)
> uhat <- Uhat(data$u)  # Observed ranks
> 
> # Plot of latent and observed data colour coded by the true component
> par(mfrow = c(1,2))
> plot(data$z, main = "Latent data", cex = 0.6,
+      xlab = "z (Experiment 1)", ylab = "z (Experiment 2)",
+      col = c("red","blue")[data$K])
> plot(uhat, main = "Observed data", cex = 0.6,
+      xlab = "u (Experiment 1)", ylab = "u (Experiment 2)",
+      col = c("red","blue")[data$K])
> 
> # Fit the model using the Pseudo EM algorithm
> init.par <- c(0.5, 1, 1, 0.5)
> res <- GMCM:::PseudoEMAlgorithm(uhat, meta2full(init.par, d = 2),
+                                 verbose = TRUE,
+                                 convergence.criterion = "absGMCM",
+                                 eps = 1e-4,
+                                 trace.theta = FALSE,
+                                 meta.special.case = TRUE)
2 | delta = 17.868577 | gmm = -2885.75 | gmcm =  167.34 
3 | delta = 8.082063 | gmm = -2907.74 | gmcm =  175.42 
4 | delta = 6.050655 | gmm = -2927.31 | gmcm =  181.47 
5 | delta = 5.348391 | gmm = -2945.36 | gmcm =  186.82 
6 | delta = 4.839607 | gmm = -2962.35 | gmcm =  191.66 
7 | delta = 4.347890 | gmm = -2978.45 | gmcm =  196.00 
8 | delta = 3.874041 | gmm = -2993.69 | gmcm =  199.88 
9 | delta = 3.435810 | gmm = -3008.08 | gmcm =  203.31 
10 | delta = 3.052414 | gmm = -3021.64 | gmcm =  206.37 
11 | delta = 2.721477 | gmm = -3034.39 | gmcm =  209.09 
12 | delta = 2.443531 | gmm = -3046.37 | gmcm =  211.53 
13 | delta = 2.211751 | gmm = -3057.62 | gmcm =  213.74 
14 | delta = 2.015994 | gmm = -3068.18 | gmcm =  215.76 
15 | delta = 1.853092 | gmm = -3078.07 | gmcm =  217.61 
16 | delta = 1.715827 | gmm = -3087.35 | gmcm =  219.33 
17 | delta = 1.599870 | gmm = -3096.02 | gmcm =  220.93 
18 | delta = 1.500794 | gmm = -3104.13 | gmcm =  222.43 
19 | delta = 1.416708 | gmm = -3111.69 | gmcm =  223.85 
20 | delta = 1.344361 | gmm = -3118.72 | gmcm =  225.19 
21 | delta = 1.282530 | gmm = -3125.24 | gmcm =  226.47 
22 | delta = 1.228756 | gmm = -3131.26 | gmcm =  227.70 
23 | delta = 1.181350 | gmm = -3136.79 | gmcm =  228.88 
24 | delta = 1.139250 | gmm = -3141.84 | gmcm =  230.02 
25 | delta = 1.100991 | gmm = -3146.39 | gmcm =  231.12 
26 | delta = 1.063712 | gmm = -3150.46 | gmcm =  232.19 
27 | delta = 1.026408 | gmm = -3154.02 | gmcm =  233.21 
28 | delta = 0.987403 | gmm = -3157.07 | gmcm =  234.20 
29 | delta = 0.942761 | gmm = -3159.60 | gmcm =  235.14 
30 | delta = 0.891736 | gmm = -3161.58 | gmcm =  236.04 
31 | delta = 0.831993 | gmm = -3163.02 | gmcm =  236.87 
32 | delta = 0.761030 | gmm = -3163.90 | gmcm =  237.63 
33 | delta = 0.680361 | gmm = -3164.23 | gmcm =  238.31 
34 | delta = 0.590134 | gmm = -3164.02 | gmcm =  238.90 
35 | delta = 0.493854 | gmm = -3163.32 | gmcm =  239.39 
36 | delta = 0.395795 | gmm = -3162.17 | gmcm =  239.79 
37 | delta = 0.301642 | gmm = -3160.62 | gmcm =  240.09 
38 | delta = 0.215391 | gmm = -3158.76 | gmcm =  240.31 
39 | delta = 0.141003 | gmm = -3156.64 | gmcm =  240.45 
40 | delta = 0.079594 | gmm = -3154.35 | gmcm =  240.53 
41 | delta = 0.031830 | gmm = -3151.94 | gmcm =  240.56 
42 | delta = 0.004377 | gmm = -3149.48 | gmcm =  240.55 
43 | delta = 0.030996 | gmm = -3147.00 | gmcm =  240.52 
44 | delta = 0.048803 | gmm = -3144.55 | gmcm =  240.47 
45 | delta = 0.061279 | gmm = -3142.14 | gmcm =  240.41 
46 | delta = 0.069590 | gmm = -3139.79 | gmcm =  240.34 
47 | delta = 0.074513 | gmm = -3137.50 | gmcm =  240.27 
48 | delta = 0.077553 | gmm = -3135.30 | gmcm =  240.19 
49 | delta = 0.079276 | gmm = -3133.16 | gmcm =  240.11 
50 | delta = 0.080348 | gmm = -3131.10 | gmcm =  240.03 
51 | delta = 0.080030 | gmm = -3129.12 | gmcm =  239.95 
52 | delta = 0.079876 | gmm = -3127.20 | gmcm =  239.87 
53 | delta = 0.079019 | gmm = -3125.35 | gmcm =  239.79 
54 | delta = 0.078102 | gmm = -3123.57 | gmcm =  239.71 
55 | delta = 0.077079 | gmm = -3121.84 | gmcm =  239.64 
56 | delta = 0.075983 | gmm = -3120.18 | gmcm =  239.56 
57 | delta = 0.074535 | gmm = -3118.57 | gmcm =  239.49 
58 | delta = 0.072803 | gmm = -3117.02 | gmcm =  239.41 
59 | delta = 0.071558 | gmm = -3115.52 | gmcm =  239.34 
60 | delta = 0.069940 | gmm = -3114.07 | gmcm =  239.27 
61 | delta = 0.068348 | gmm = -3112.66 | gmcm =  239.20 
62 | delta = 0.066729 | gmm = -3111.31 | gmcm =  239.14 
63 | delta = 0.065271 | gmm = -3110.00 | gmcm =  239.07 
64 | delta = 0.063689 | gmm = -3108.74 | gmcm =  239.01 
65 | delta = 0.061805 | gmm = -3107.52 | gmcm =  238.95 
66 | delta = 0.060516 | gmm = -3106.34 | gmcm =  238.89 
67 | delta = 0.058488 | gmm = -3105.20 | gmcm =  238.83 
68 | delta = 0.057113 | gmm = -3104.10 | gmcm =  238.77 
69 | delta = 0.055391 | gmm = -3103.04 | gmcm =  238.72 
70 | delta = 0.053966 | gmm = -3102.01 | gmcm =  238.66 
71 | delta = 0.052288 | gmm = -3101.02 | gmcm =  238.61 
72 | delta = 0.050634 | gmm = -3100.07 | gmcm =  238.56 
73 | delta = 0.049541 | gmm = -3099.15 | gmcm =  238.51 
74 | delta = 0.047722 | gmm = -3098.26 | gmcm =  238.46 
75 | delta = 0.046379 | gmm = -3097.40 | gmcm =  238.41 
76 | delta = 0.044991 | gmm = -3096.57 | gmcm =  238.37 
77 | delta = 0.043599 | gmm = -3095.78 | gmcm =  238.33 
78 | delta = 0.042113 | gmm = -3095.01 | gmcm =  238.28 
79 | delta = 0.040781 | gmm = -3094.27 | gmcm =  238.24 
80 | delta = 0.039585 | gmm = -3093.55 | gmcm =  238.20 
81 | delta = 0.038263 | gmm = -3092.87 | gmcm =  238.17 
82 | delta = 0.037011 | gmm = -3092.20 | gmcm =  238.13 
83 | delta = 0.035897 | gmm = -3091.56 | gmcm =  238.09 
84 | delta = 0.034441 | gmm = -3090.95 | gmcm =  238.06 
85 | delta = 0.033345 | gmm = -3090.36 | gmcm =  238.02 
86 | delta = 0.032295 | gmm = -3089.78 | gmcm =  237.99 
87 | delta = 0.031258 | gmm = -3089.23 | gmcm =  237.96 
88 | delta = 0.030116 | gmm = -3088.71 | gmcm =  237.93 
89 | delta = 0.029077 | gmm = -3088.20 | gmcm =  237.90 
90 | delta = 0.028054 | gmm = -3087.70 | gmcm =  237.87 
91 | delta = 0.027176 | gmm = -3087.23 | gmcm =  237.85 
92 | delta = 0.026403 | gmm = -3086.78 | gmcm =  237.82 
93 | delta = 0.025197 | gmm = -3086.34 | gmcm =  237.80 
94 | delta = 0.024321 | gmm = -3085.92 | gmcm =  237.77 
95 | delta = 0.023404 | gmm = -3085.51 | gmcm =  237.75 
96 | delta = 0.022557 | gmm = -3085.12 | gmcm =  237.72 
97 | delta = 0.021739 | gmm = -3084.75 | gmcm =  237.70 
98 | delta = 0.021174 | gmm = -3084.38 | gmcm =  237.68 
99 | delta = 0.020324 | gmm = -3084.04 | gmcm =  237.66 
100 | delta = 0.019571 | gmm = -3083.70 | gmcm =  237.64 
101 | delta = 0.018864 | gmm = -3083.38 | gmcm =  237.62 
102 | delta = 0.018122 | gmm = -3083.07 | gmcm =  237.60 
103 | delta = 0.017433 | gmm = -3082.78 | gmcm =  237.59 
104 | delta = 0.016851 | gmm = -3082.49 | gmcm =  237.57 
105 | delta = 0.016232 | gmm = -3082.21 | gmcm =  237.55 
106 | delta = 0.015583 | gmm = -3081.95 | gmcm =  237.54 
107 | delta = 0.014975 | gmm = -3081.69 | gmcm =  237.52 
108 | delta = 0.014420 | gmm = -3081.45 | gmcm =  237.51 
109 | delta = 0.013892 | gmm = -3081.22 | gmcm =  237.50 
110 | delta = 0.013395 | gmm = -3080.99 | gmcm =  237.48 
111 | delta = 0.012932 | gmm = -3080.77 | gmcm =  237.47 
112 | delta = 0.012513 | gmm = -3080.56 | gmcm =  237.46 
113 | delta = 0.012074 | gmm = -3080.36 | gmcm =  237.44 
114 | delta = 0.011569 | gmm = -3080.17 | gmcm =  237.43 
115 | delta = 0.011113 | gmm = -3079.98 | gmcm =  237.42 
116 | delta = 0.010670 | gmm = -3079.80 | gmcm =  237.41 
117 | delta = 0.010217 | gmm = -3079.63 | gmcm =  237.40 
118 | delta = 0.009841 | gmm = -3079.47 | gmcm =  237.39 
119 | delta = 0.009477 | gmm = -3079.31 | gmcm =  237.38 
120 | delta = 0.009106 | gmm = -3079.16 | gmcm =  237.37 
121 | delta = 0.008760 | gmm = -3079.01 | gmcm =  237.36 
122 | delta = 0.008448 | gmm = -3078.87 | gmcm =  237.36 
123 | delta = 0.008137 | gmm = -3078.73 | gmcm =  237.35 
124 | delta = 0.007861 | gmm = -3078.60 | gmcm =  237.34 
125 | delta = 0.007550 | gmm = -3078.48 | gmcm =  237.33 
126 | delta = 0.007260 | gmm = -3078.36 | gmcm =  237.32 
127 | delta = 0.007005 | gmm = -3078.24 | gmcm =  237.32 
128 | delta = 0.006747 | gmm = -3078.13 | gmcm =  237.31 
129 | delta = 0.006459 | gmm = -3078.03 | gmcm =  237.30 
130 | delta = 0.006158 | gmm = -3077.92 | gmcm =  237.30 
131 | delta = 0.005900 | gmm = -3077.83 | gmcm =  237.29 
132 | delta = 0.005662 | gmm = -3077.73 | gmcm =  237.29 
133 | delta = 0.005437 | gmm = -3077.64 | gmcm =  237.28 
134 | delta = 0.005224 | gmm = -3077.55 | gmcm =  237.28 
135 | delta = 0.005026 | gmm = -3077.47 | gmcm =  237.27 
136 | delta = 0.004835 | gmm = -3077.39 | gmcm =  237.27 
137 | delta = 0.004638 | gmm = -3077.31 | gmcm =  237.26 
138 | delta = 0.004463 | gmm = -3077.24 | gmcm =  237.26 
139 | delta = 0.004294 | gmm = -3077.17 | gmcm =  237.25 
140 | delta = 0.004129 | gmm = -3077.10 | gmcm =  237.25 
141 | delta = 0.003973 | gmm = -3077.03 | gmcm =  237.24 
142 | delta = 0.003827 | gmm = -3076.97 | gmcm =  237.24 
143 | delta = 0.003688 | gmm = -3076.91 | gmcm =  237.24 
144 | delta = 0.003561 | gmm = -3076.85 | gmcm =  237.23 
145 | delta = 0.003434 | gmm = -3076.79 | gmcm =  237.23 
146 | delta = 0.003304 | gmm = -3076.74 | gmcm =  237.23 
147 | delta = 0.003184 | gmm = -3076.69 | gmcm =  237.22 
148 | delta = 0.003074 | gmm = -3076.64 | gmcm =  237.22 
149 | delta = 0.002949 | gmm = -3076.59 | gmcm =  237.22 
150 | delta = 0.002832 | gmm = -3076.54 | gmcm =  237.21 
151 | delta = 0.002717 | gmm = -3076.50 | gmcm =  237.21 
152 | delta = 0.002608 | gmm = -3076.46 | gmcm =  237.21 
153 | delta = 0.002507 | gmm = -3076.41 | gmcm =  237.21 
154 | delta = 0.002412 | gmm = -3076.38 | gmcm =  237.20 
155 | delta = 0.002342 | gmm = -3076.34 | gmcm =  237.20 
156 | delta = 0.002207 | gmm = -3076.30 | gmcm =  237.20 
157 | delta = 0.002119 | gmm = -3076.27 | gmcm =  237.20 
158 | delta = 0.002034 | gmm = -3076.23 | gmcm =  237.20 
159 | delta = 0.001951 | gmm = -3076.20 | gmcm =  237.19 
160 | delta = 0.001873 | gmm = -3076.17 | gmcm =  237.19 
161 | delta = 0.001798 | gmm = -3076.14 | gmcm =  237.19 
162 | delta = 0.001727 | gmm = -3076.11 | gmcm =  237.19 
163 | delta = 0.001658 | gmm = -3076.08 | gmcm =  237.19 
164 | delta = 0.001592 | gmm = -3076.06 | gmcm =  237.19 
165 | delta = 0.001529 | gmm = -3076.03 | gmcm =  237.18 
166 | delta = 0.001467 | gmm = -3076.01 | gmcm =  237.18 
167 | delta = 0.001409 | gmm = -3075.99 | gmcm =  237.18 
168 | delta = 0.001353 | gmm = -3075.96 | gmcm =  237.18 
169 | delta = 0.001299 | gmm = -3075.94 | gmcm =  237.18 
170 | delta = 0.001248 | gmm = -3075.92 | gmcm =  237.18 
171 | delta = 0.001199 | gmm = -3075.90 | gmcm =  237.18 
172 | delta = 0.001152 | gmm = -3075.88 | gmcm =  237.17 
173 | delta = 0.001107 | gmm = -3075.86 | gmcm =  237.17 
174 | delta = 0.001063 | gmm = -3075.85 | gmcm =  237.17 
175 | delta = 0.001021 | gmm = -3075.83 | gmcm =  237.17 
176 | delta = 0.000980 | gmm = -3075.81 | gmcm =  237.17 
177 | delta = 0.000942 | gmm = -3075.80 | gmcm =  237.17 
178 | delta = 0.000904 | gmm = -3075.78 | gmcm =  237.17 
179 | delta = 0.000869 | gmm = -3075.77 | gmcm =  237.17 
180 | delta = 0.000834 | gmm = -3075.76 | gmcm =  237.17 
181 | delta = 0.000802 | gmm = -3075.74 | gmcm =  237.17 
182 | delta = 0.000771 | gmm = -3075.73 | gmcm =  237.17 
183 | delta = 0.000741 | gmm = -3075.72 | gmcm =  237.16 
184 | delta = 0.000712 | gmm = -3075.71 | gmcm =  237.16 
185 | delta = 0.000684 | gmm = -3075.70 | gmcm =  237.16 
186 | delta = 0.000657 | gmm = -3075.68 | gmcm =  237.16 
187 | delta = 0.000633 | gmm = -3075.67 | gmcm =  237.16 
188 | delta = 0.000608 | gmm = -3075.66 | gmcm =  237.16 
189 | delta = 0.000585 | gmm = -3075.65 | gmcm =  237.16 
190 | delta = 0.000562 | gmm = -3075.65 | gmcm =  237.16 
191 | delta = 0.000540 | gmm = -3075.64 | gmcm =  237.16 
192 | delta = 0.000520 | gmm = -3075.63 | gmcm =  237.16 
193 | delta = 0.000499 | gmm = -3075.62 | gmcm =  237.16 
194 | delta = 0.000480 | gmm = -3075.61 | gmcm =  237.16 
195 | delta = 0.000462 | gmm = -3075.60 | gmcm =  237.16 
196 | delta = 0.000444 | gmm = -3075.60 | gmcm =  237.16 
197 | delta = 0.000427 | gmm = -3075.59 | gmcm =  237.16 
198 | delta = 0.000411 | gmm = -3075.58 | gmcm =  237.16 
199 | delta = 0.000395 | gmm = -3075.58 | gmcm =  237.16 
200 | delta = 0.000380 | gmm = -3075.57 | gmcm =  237.16 
201 | delta = 0.000365 | gmm = -3075.57 | gmcm =  237.16 
202 | delta = 0.000351 | gmm = -3075.56 | gmcm =  237.15 
203 | delta = 0.000337 | gmm = -3075.55 | gmcm =  237.15 
204 | delta = 0.000324 | gmm = -3075.55 | gmcm =  237.15 
205 | delta = 0.000312 | gmm = -3075.54 | gmcm =  237.15 
206 | delta = 0.000300 | gmm = -3075.54 | gmcm =  237.15 
207 | delta = 0.000288 | gmm = -3075.53 | gmcm =  237.15 
208 | delta = 0.000277 | gmm = -3075.53 | gmcm =  237.15 
209 | delta = 0.000266 | gmm = -3075.53 | gmcm =  237.15 
210 | delta = 0.000256 | gmm = -3075.52 | gmcm =  237.15 
211 | delta = 0.000246 | gmm = -3075.52 | gmcm =  237.15 
212 | delta = 0.000236 | gmm = -3075.51 | gmcm =  237.15 
213 | delta = 0.000227 | gmm = -3075.51 | gmcm =  237.15 
214 | delta = 0.000218 | gmm = -3075.51 | gmcm =  237.15 
215 | delta = 0.000210 | gmm = -3075.50 | gmcm =  237.15 
216 | delta = 0.000202 | gmm = -3075.50 | gmcm =  237.15 
217 | delta = 0.000194 | gmm = -3075.50 | gmcm =  237.15 
218 | delta = 0.000186 | gmm = -3075.49 | gmcm =  237.15 
219 | delta = 0.000179 | gmm = -3075.49 | gmcm =  237.15 
220 | delta = 0.000172 | gmm = -3075.49 | gmcm =  237.15 
221 | delta = 0.000165 | gmm = -3075.48 | gmcm =  237.15 
222 | delta = 0.000159 | gmm = -3075.48 | gmcm =  237.15 
223 | delta = 0.000153 | gmm = -3075.48 | gmcm =  237.15 
224 | delta = 0.000147 | gmm = -3075.48 | gmcm =  237.15 
225 | delta = 0.000141 | gmm = -3075.48 | gmcm =  237.15 
226 | delta = 0.000135 | gmm = -3075.47 | gmcm =  237.15 
227 | delta = 0.000130 | gmm = -3075.47 | gmcm =  237.15 
228 | delta = 0.000125 | gmm = -3075.47 | gmcm =  237.15 
229 | delta = 0.000120 | gmm = -3075.47 | gmcm =  237.15 
230 | delta = 0.000116 | gmm = -3075.46 | gmcm =  237.15 
231 | delta = 0.000111 | gmm = -3075.46 | gmcm =  237.15 
232 | delta = 0.000107 | gmm = -3075.46 | gmcm =  237.15 
233 | delta = 0.000103 | gmm = -3075.46 | gmcm =  237.15 
234 | delta = 0.000099 | gmm = -3075.46 | gmcm =  237.15 
> 
> # Compute posterior cluster probabilities
> IDRs <- get.IDR(uhat, par = full2meta(res$theta))
> 
> # Plot of observed data colour coded by the MAP estimate
> plot(res$loglik[3,], main = "Loglikelihood trace", type = "l",
+      ylab = "log GMCM likelihood")
> abline(v = which.max(res$loglik[3,])) # Chosen MLE
> plot(uhat, main = "Clustering\nIDR < 0.05", xlab = "", ylab = "", cex = 0.6,
+      col = c("Red","Blue")[IDRs$Khat])
> 
> # View parameters
> rbind(init.par, true.par, estimate = full2meta(res$theta))
              pie1    mu.mu     sigma       rho
init.par 0.5000000 1.000000 1.0000000 0.5000000
true.par 0.8000000 3.000000 1.5000000 0.4000000
estimate 0.8224538 2.521749 0.8548858 0.4069159
> 
> # Confusion matrix
> table("Khat" = IDRs$Khat, "K" = data$K)
    K
Khat   1   2
   1 797  37
   2   5 161
> 
> 
> 
> 
> 
> dev.off()
null device 
          1 
>


Created & Maintained by Osamu Ogasawara (osamu.ogasawara@gmail.com) and IMSBIO