need help about "lavaan" and "blavaan"

luquany...@gmail.com

unread,

May 4, 2016, 3:43:00 AM5/4/16

to lavaan

I am a beginner in lavaan and blavaan library. I have some basic questions in my code.

library(lavaan)

x<-read.csv("C:/Users/Quanying LU/Desktop/5.csv")

co2model<-'"

> library(lavaan)

> x<-read.csv("C:/Users/Quanying LU/Desktop/5.csv")

> co2model<-'

+ # measurement model

+ co2=~y1

+ total=~y2+y3+y4+y5+y6

+ struct=~y7+y8+y9+y10

+ tech=~y11

+ # regressions

+ co2~total+struct+tech

+ total~struct+tech

+ struct~tech

+ # residual covariances

+ y2~~y3+y4+y5+y6+y7+y8+y9+y10+y11

+ y3 ~~ y4+y7

+ y5 ~~ y4+y7+y8+y9+y11

+ y6 ~~ y4+y7+y8+y9+y11

+ y7~~y4

+ y8~~y11

+ y9~~y11

+ y10~~y11'

> fit <- sem(co2model, data = x,meanstructure = "default",

+ conditional.x = "default", fixed.x = "default",

+ orthogonal = FALSE, std.lv = FALSE,

+ parameterization = "default", std.ov = FALSE,

+ missing = "default", ordered = NULL,

+ sample.cov = NULL, sample.cov.rescale = "default",

+ sample.mean = NULL, sample.nobs = NULL,

+ ridge = 1e-05, group = NULL,

+ group.label = NULL, group.equal = "", group.partial = "",

+ group.w.free = FALSE, cluster = NULL, constraints = '',

+ estimator = "default", likelihood = "default", link = "default",

+ information = "default", se = "default", test = "default",

+ bootstrap = 1000L, mimic = "default", representation = "default",

+ do.fit = TRUE, control = list(), WLS.V = NULL, NACOV = NULL,

+ zero.add = "default", zero.keep.margins = "default",zero.cell.warn = TRUE,

+ start = "default", verbose = FALSE, warn = TRUE, debug = FALSE)

Warning messages:

1: In lav_object_post_check(lavobject) :

  lavaan WARNING: some estimated variances are negative

2: In lav_object_post_check(lavobject) :

  lavaan WARNING: covariance matrix of latent variables is not positive definite; use inspect(fit,"cov.lv") to investigate.

3: In lav_object_post_check(lavobject) :

  lavaan WARNING: observed variable error term matrix (theta) is not positive definite; use inspect(fit,"theta") to investigate.

> summary(fit, standardized = TRUE)

lavaan (0.5-20) converged normally after 619 iterations

  Number of observations                            27

  Estimator                                         ML

  Minimum Function Test Statistic               86.803

  Degrees of freedom                                15

  P-value (Chi-square)                           0.000

Parameter Estimates:

  Information                                 Expected

  Standard Errors                             Standard

Latent Variables:

                   Estimate  Std.Err  Z-value  P(>|z|)   Std.lv  Std.all

  co2 =~

    y1                1.000                               0.863    1.000

  total =~

    y2                1.000                               0.736    0.969

    y3                0.094    0.003   28.857    0.000    0.069    0.974

    y4                1.539    0.048   32.027    0.000    1.132    0.966

    y5                1.043    0.088   11.861    0.000    0.767    0.905

    y6                0.652    0.040   16.395    0.000    0.480    0.955

  struct =~

    y7                1.000                               0.243    0.999

    y8                0.196    0.054    3.647    0.000    0.048    0.544

    y9                2.795    0.076   36.652    0.000    0.679    0.991

    y10              -1.052    0.065  -16.125    0.000   -0.256   -0.952

  tech =~

    y11               1.000                               0.307    1.000

Regressions:

                   Estimate  Std.Err  Z-value  P(>|z|)   Std.lv  Std.all

  co2 ~

    total             0.864    0.190    4.547    0.000    0.737    0.737

    struct            0.910    0.817    1.113    0.266    0.256    0.256

    tech              0.032    0.309    0.104    0.917    0.011    0.011

  total ~

    struct            3.043    0.452    6.731    0.000    1.005    1.005

    tech             -0.043    0.347   -0.125    0.900   -0.018   -0.018

  struct ~

    tech             -0.766    0.037  -20.437    0.000   -0.969   -0.969

Covariances:

                   Estimate  Std.Err  Z-value  P(>|z|)   Std.lv  Std.all

  y2 ~~

    y3                0.002    0.001    3.731    0.000    0.002    0.728

    y4                0.046    0.010    4.625    0.000    0.046    0.808

    y5                0.024    0.006    4.076    0.000    0.024    0.353

    y6                0.010    0.002    4.249    0.000    0.010    0.347

    y7               -0.005    0.001   -4.376    0.000   -0.005   -2.480

    y8               -0.001    0.001   -1.003    0.316   -0.001   -0.051

    y9               -0.006    0.002   -3.769    0.000   -0.006   -0.349

    y10              -0.007    0.001   -4.367    0.000   -0.007   -0.420

    y11               0.003    0.001    2.483    0.013    0.003      Inf

  y3 ~~

    y4                0.004    0.001    3.971    0.000    0.004    0.849

    y7               -0.001    0.000   -4.915    0.000   -0.001   -3.000

  y4 ~~

    y5                0.034    0.010    3.439    0.001    0.034    0.313

  y5 ~~

    y7               -0.004    0.002   -2.030    0.042   -0.004   -0.870

    y8                0.011    0.003    3.242    0.001    0.011    0.425

    y9               -0.017    0.005   -3.619    0.000   -0.017   -0.501

    y11               0.002    0.004    0.499    0.618    0.002      Inf

  y4 ~~

    y6                0.012    0.003    3.477    0.001    0.012    0.261

  y6 ~~

    y7               -0.003    0.001   -3.632    0.000   -0.003   -2.001

    y8               -0.008    0.002   -3.785    0.000   -0.008   -0.744

    y9               -0.009    0.002   -3.502    0.000   -0.009   -0.634

    y11               0.011    0.003    3.896    0.000    0.011      Inf

  y4 ~~

    y7               -0.008    0.002   -4.256    0.000   -0.008   -2.301

  y8 ~~

    y11              -0.003    0.001   -2.990    0.003   -0.003     -Inf

  y9 ~~

    y11              -0.003    0.001   -2.609    0.009   -0.003     -Inf

  y10 ~~

    y11               0.001    0.001    0.855    0.393    0.001      Inf

Variances:

                   Estimate  Std.Err  Z-value  P(>|z|)   Std.lv  Std.all

    y1                0.000                               0.000    0.000

    y2                0.036    0.008    4.713    0.000    0.036    0.062

    y3                0.000    0.000    3.938    0.000    0.000    0.052

    y4                0.093    0.018    5.049    0.000    0.093    0.067

    y5                0.130    0.032    4.071    0.000    0.130    0.181

    y6                0.022    0.006    3.705    0.000    0.022    0.087

    y7                0.000    0.000    1.263    0.206    0.000    0.002

    y8                0.005    0.001    4.198    0.000    0.005    0.704

    y9                0.008    0.002    4.790    0.000    0.008    0.018

    y10               0.007    0.001    4.791    0.000    0.007    0.094

    y11               0.000                               0.000    0.000

    co2               0.021    0.006    3.381    0.001    0.028    0.028

    total            -0.025    0.007   -3.391    0.001   -0.045   -0.045

    struct            0.004    0.001    3.943    0.000    0.061    0.061

    tech              0.094    0.026    3.685    0.000    1.000    1.000

library(blavaan)

library(rjags)

library(runjags)

x<-read.csv("C:/Users/Quanying LU/Desktop/5.csv")

co2.model<-'

# measurement model

co2=~y1

total=~y2+y3+y4+y5+y6

struct=~y7+y8+y9+y10

tech=~y11

# regressions

co2~total+struct+tech

total~struct+tech

struct~tech

# residual covariances

y2~~y3+y4+y5+y6+y7+y8+y9+y10+y11

y3 ~~ y4+y7

y5 ~~ y7+y8+y9

y6 ~~ y7+y8+y9+y11

y7~~y8

y8~~y11

y10~~y11'

fit <- bsem(co2.model, data=x,

dp=dpriors(nu="dnorm(5,1e-3)", itheta="dlnorm(1,.1)[sd]",

ipsi="dlnorm(1,.1)[sd]", rho="dbeta(3,3)"),

jagcontrol=list(method="rjparallel"))

summary(fit)

Failed with error: ‘there is no package called ‘modeest’’

Compiling rjags model...

Error : The following error occured when compiling and adapting the model using rjags:

Error in rjags::jags.model(model, data = dataenv, inits = inits, n.chains = length(runjags.object$end.state), :

Error in node theta[1,1]

Invalid parent values

It may help to use failed.jags(c('model','data','inits')) to see model/data/inits syntax with line numbers

Error in blavaan(co2.model, data = x, dp = dpriors(nu = "dnorm(5,1e-3)", :

blavaan ERROR: problem with jags estimation. The jags model and data have been exported.

luquany...@gmail.com

unread,

May 4, 2016, 3:44:11 AM5/4/16

to lavaan

I think that there may be some mistakes in my "residual covariances" setting. I do not know how to set this part in all codes.

在 2016年5月4日星期三 UTC+8下午3:43:00，luquany...@gmail.com写道：

Terrence Jorgensen

unread,

May 4, 2016, 4:06:29 AM5/4/16

to lavaan

You are using the correct lavaan syntax (~~) to specify residual covariances, assuming you have informative enough priors to empirically identify this model. There is actually a blavaan Google group where you can post questions about that package.

https://groups.google.com/d/forum/blavaan

Terrence D. Jorgensen

Postdoctoral Researcher, Methods and Statistics

Research Institute for Child Development and Education, the University of Amsterdam

UvA web page: http://www.uva.nl/profile/t.d.jorgensen

luquany...@gmail.com

unread,

May 4, 2016, 5:09:36 AM5/4/16

to lavaan

Thank you so much!

在 2016年5月4日星期三 UTC+8下午4:06:29，Terrence Jorgensen写道：

luquany...@gmail.com

unread,

May 5, 2016, 5:41:39 AM5/5/16

to lavaan

Dear Terrence Jorgensen:

Do you know how to set residual covariances in the model?

在 2016年5月4日星期三 UTC+8下午5:09:36，luquany...@gmail.com写道：

Reply all

Reply to author

Forward