Diagnostics of Mixture Objects Fitted with 'mle2'

Return diagnostics after on a mixture distribution object fittted with 'mx_metafit'

diagnostics(obj, ...)

# S4 method for mxMle
diagnostics(obj, breaks = NULL, counts = NULL, ...)

Arguments

obj

object of class mxMle
...

reserved for future extensions
breaks

class boundaries of the data
counts

frequency of observations

Value

list of diagnostic measures

Examples


breaks <- 0:28
counts <- c(36, 0, 2, 3, 4, 8, 9, 14, 10, 9, 3, 1, 1, 2,
            4, 8, 20, 45, 40, 54, 41, 22, 8, 3, 3, 0, 0,0)

observations <- unbin(breaks[-1], counts) # upper class boundaries

(comp <- mx_guess_components(observations, bw=2/3, mincut=0.9))

#>        mean        sd         L
#> 1  1.040054 0.7032986 0.1055751
#> 2  7.805322 2.0509038 0.1775396
#> 3 19.505968 2.1153693 0.7168853
obj <- mxObj(comp, left="e")
ret <- mx_metafit(breaks, counts, obj)

summary(ret)       # general summary

#> Maximum likelihood estimation
#> 
#> Call:
#> mle2(minuslogl = llunimix, start = parms)
#> 
#> Coefficients:
#>        Estimate Std. Error  z value     Pr(z)    
#> L1     0.102637   0.016246   6.3179 2.652e-10 ***
#> L2     0.183466   0.020879   8.7872 < 2.2e-16 ***
#> rate1  7.891433   8.638539   0.9135     0.361    
#> mean2  7.304443   0.283240  25.7889 < 2.2e-16 ***
#> sd2    2.140047   0.226415   9.4519 < 2.2e-16 ***
#> mean3 19.029089   0.126390 150.5588 < 2.2e-16 ***
#> sd3    1.948262   0.093759  20.7795 < 2.2e-16 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> -2 log L: 1873.47 
cov2cor(vcov(ret)) # correlation matrix of parameters

#>                  L1            L2         rate1         mean2           sd2
#> L1     1.000000e+00 -1.614179e-01  1.899205e-05  4.404988e-03 -0.0129406201
#> L2    -1.614179e-01  1.000000e+00 -2.322614e-05  3.204188e-02  0.0689235221
#> rate1  1.899205e-05 -2.322614e-05  1.000000e+00  3.896354e-05 -0.0001705144
#> mean2  4.404988e-03  3.204188e-02  3.896354e-05  1.000000e+00  0.1155702129
#> sd2   -1.294062e-02  6.892352e-02 -1.705144e-04  1.155702e-01  1.0000000000
#> mean3 -7.909409e-04  2.135825e-02 -1.098725e-05  4.706817e-02  0.0768120135
#> sd3    1.378197e-03 -3.868433e-02  1.916903e-05 -8.413941e-02 -0.1346087859
#>               mean3           sd3
#> L1    -7.909409e-04  1.378197e-03
#> L2     2.135825e-02 -3.868433e-02
#> rate1 -1.098725e-05  1.916903e-05
#> mean2  4.706817e-02 -8.413941e-02
#> sd2    7.681201e-02 -1.346088e-01
#> mean3  1.000000e+00 -5.204826e-02
#> sd3   -5.204826e-02  1.000000e+00
diagnostics(ret)   # additional diagnostics

#> 
#> Convergence =  0 :  Successful completion of optim 
#> 
#> Agreement of (cumulative) distribution to data:
#> 
#> cor(observed, predicted)):          0.9998638 
#> rsquared (explained variance):      0.9997274 
#> Nash Sutcliffe Efficiency:          0.9997274 
#> Log Likelihood:                     -936.735 
#> AIC:                                1887.47 
#> Total number of counts:             350 
#> Number of non-empty classes:        24 
#>