Message-ID: <503F7CDE.7040308@yeah.net>
Date: 2012-08-30T14:46:54Z
From: Jinsong Zhao
Subject: 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