Dear Jinsong,
This model is grossly underidentified because there are no exogenous
variables in it. Your inability to estimate the model isn't a software
issue.
Best,
John
-----------------------------------------------
John Fox
Senator McMaster Professor of Social Statistics
Department of Sociology
McMaster University
Hamilton, Ontario, Canada
> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at
r-project.org]
> On Behalf Of Jinsong Zhao
> Sent: Thursday, August 30, 2012 10:47 AM
> To: r-help at r-project.org
> Subject: [R] path analysis help
>
> Hi there,
>
> I searched R-help list with "path analysis" as keyword, and learn
that
> sem package can do it. However, I don't figure out a way to construct
> the model for the path diagram as Fig. 1. in Huang et al. (2002)[1].
>
> I try the following code:
>
> huang.cor <- readMoments(diag=FALSE, names=c('x1', 'x2',
'x3', 'y'))
> 0.76
> 0.91 0.72
> 0.94 0.77 0.83
>
> huang.mod <- specifyModel()
> x1 -> y, p1
> x2 -> y, p2
> x3 -> y, p3
> x1 -> x2, p12
> x2 -> x1, p21
> x2 -> x3, p23
> x3 -> x2, p32
> x1 -> x3, p13
> x3 -> x1, p31
>
> huang.sem <- sem(huang.mod, huang.cor, 100)# 100 is arbitarious.
>
> It give the error message:
>
> Error in sem.default(ram, S = S, N = N, raw = raw, data = data,
> param.names = pars, :
> The model has negative degrees of freedom = -3
>
> I don't know why.
>
> I hope to get direct effect, indirect effect and total effects for every
> variable. However, I don't figure out how to do with sem package. It
> there any other package that can do it?
>
> Any suggestion or hint will be greatly appreciated.
>
> Regards,
> Jinsong
>
>
> [1] Huang, B., Thornhill, N., Shah, S. and Shook, D. (2002). Path
> analysis for process troubleshooting. Proceedings of Advanced Control of
> Industrial Processes, Kumamoto, Japan, 10-12 June, 149-154.
> url: http://eprints.ucl.ac.uk/494/1/HuangEtAl_AdConIP_2002.pdf
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.