code

This is code for bi-variate RI-CLPM

RICLPM <- '
a1~mua1*1 #fix the intercept of a1 to mean of a1, which is a freely estimated parameter 
m1~mum1*1 #this step already defines the latent variables
a2~mua2*1 
m2~mum2*1 
a3~mua3*1 
m3~mum3*1 
a4~mua4*1 
m4~mum4*1  
a5~mua5*1 
m5~mum5*1 
a6~mua6*1 
m6~mum6*1
a7~mua7*1
m7~mum7*1


a1~~0*a1 #variance of a1 is 0
m1~~0*m1 
a2~~0*a2 
m2~~0*m2 
a3~~0*a3 
m3~~0*m3 
a4~~0*a4 
m4~~0*m4 
a5~~0*a5 
m5~~0*m5 
a6~~0*a6 
m6~~0*m6
a7~~0*a7 
m7~~0*m7


Ia~~0*FFa1 # covariance of Ia and FFa1 is fixed to 0 -> Ia, Im, FFa1, and FFm1 are orthogonal factors
Ia~~0*FFm1 
Im~~0*FFa1 
Im~~0*FFm1

Ia=~1*a1+1*a2+1*a3+1*a4+1*a5+1*a6+1*a7 #define Ia 
Im=~1*m1+1*m2+1*m3+1*m4+1*m5+1*m6+1*m7  #define Im 

Ia~0*1 #the intercept of Ia is 0?? 
Im~0*1 #the intercept of Im is 0?? 
Ia~~taua*Ia #variance of Ia is taux 
Im~~taum*Im #variance of Im is tauy
Ia~~tauam*Im #covariance of Ia and Im is tauam

FFa1~0*1 #the intercept of FFa1 is 0 
FFm1~0*1 #the intercept of FFm1 is 0 
FFa1~~phia*FFa1 #variance of FFa1 is phia
FFm1~~phim*FFm1 #variance of FFm1 is phim
FFa1~~phiam*FFm1 #covariance of FFa1 and FFm1 is phiam (yellow curved arrow)

#FFa1~ gammaa*FFm1 # 24-03-25: found this line, it doesnt exist for all models, but it doesnt matter, the results stay the same with or without it 

FFm2~ betam*FFm1+gammam*FFa1 
FFa2~ betaa*FFa1+gammaa*FFm2

FFm3~ betam*FFm2+gammam*FFa2 
FFa3~ betaa*FFa2+gammaa*FFm3

FFm4~ betam*FFm3+gammam*FFa3 
FFa4~ betaa*FFa3+gammaa*FFm4

FFm5~ betam*FFm4+gammam*FFa4 
FFa5~ betaa*FFa4+gammaa*FFm5

FFm6~ betam*FFm5+gammam*FFa5 
FFa6~ betaa*FFa5+gammaa*FFm6

FFm7~ betam*FFm6+gammam*FFa6 
FFa7~ betaa*FFa6+gammaa*FFm7


FFa2~~Omegaa*FFa2 #variance of FFa2 is Omegaa
FFa3~~Omegaa*FFa3 
FFa4~~Omegaa*FFa4 
FFa5~~Omegaa*FFa5 
FFa6~~Omegaa*FFa6
FFa7~~Omegaa*FFa7


FFm2~~Omegam*FFm2 #variance of FFm2 is Omegam
FFm3~~Omegam*FFm3 
FFm4~~Omegam*FFm4 
FFm5~~Omegam*FFm5 
FFm6~~Omegam*FFm6
FFm7~~Omegam*FFm7


FFa2~~Omegaam*FFm2 #covariance of FFa2 and FFm2 is Omegaam
FFa3~~Omegaam*FFm3 # 24-03-15: Abe and I found out somehow we skipped these & added them again
FFa4~~Omegaam*FFm4 
FFa5~~Omegaam*FFm5
FFa6~~Omegaam*FFm6
FFa7~~Omegaam*FFm7


FFm1 =~ 1*m1 #define the latent variable f*m1 (from fm) #here Abe suggests to also freely estimate insteading of fixing the parameters to 1
FFm2 =~ 1*m2 
FFm3 =~ 1*m3 
FFm4 =~ 1*m4 
FFm5 =~ 1*m5 
FFm6 =~ 1*m6
FFm7 =~ 1*m7


FFa1 =~ 1*a1 #define the latent variable a*1 (from fa)
FFa2 =~ 1*a2 
FFa3 =~ 1*a3 
FFa4 =~ 1*a4 
FFa5 =~ 1*a5 
FFa6 =~ 1*a6
FFa7 =~ 1*a7

'

#this is for the tri-variate model

RICLPM_tri <- '
a1~mua1*1 #fix the intercept of a1 to mean of a1, which is a freely estimated parameter 
m1~mum1*1 #this step already defines the latent variableS
t1~mut1*1
a2~mua2*1 
m2~mum2*1 
t2~mut2*1 
a3~mua3*1 
m3~mum3*1 
t3~mut3*1 
a4~mua4*1 
m4~mum4*1  
t4~mut4*1
a5~mua5*1 
m5~mum5*1 
t5~mut5*1 
a6~mua6*1 
m6~mum6*1 
t6~mut6*1 
a7~mua7*1 
m7~mum7*1 
t7~mut7*1 


a1~~0*a1 #variance of a1 is 0
m1~~0*m1 
t1~~0*t1 
a2~~0*a2 
m2~~0*m2 
t2~~0*t2 
a3~~0*a3 
m3~~0*m3 
t3~~0*t3 
a4~~0*a4 
m4~~0*m4 
t4~~0*t4 
a5~~0*a5 
m5~~0*m5 
t5~~0*t5 
a6~~0*a6 
m6~~0*m6
t6~~0*t6
a7~~0*a7 
m7~~0*m7 
t7~~0*t7 


Ia~~0*FFa1 # covariance of Ia and FFa1 is fixed to 0 -> Ia, Im, FFa1, and FFm1 are orthogonal factors
Ia~~0*FFm1 
Ia~~0*FFt1
Im~~0*FFa1 
Im~~0*FFm1
Im~~0*FFt1
It~~0*FFm1
It~~0*FFa1
It~~0*FFt1

Ia=~1*a1+1*a2+1*a3+1*a4+1*a5+1*a6+1*a7 #define Ia 
Im=~1*m1+1*m2+1*m3+1*m4+1*m5+1*m6+1*m7 #define Im 
It=~1*t1+1*t2+1*t3+1*t4+1*t5+1*t6+1*t7 #define It 

Ia~0*1 #the intercept of Ia is 0?? 
Im~0*1 #the intercept of Im is 0?? 
It~0*1
Ia~~taua*Ia #variance of Ia is taux 
Im~~taum*Im #variance of Im is tauy
It~~taut*It #variance of It is taut

Ia~~tauat*It #covariance of Ia and It is tauat
Im~~taumt*It #covariance of Im and It is taumt
Ia~~tauam*Im #covariance of Im and Ia is tauam

FFa1~0*1 #the intercept of FFa1 is 0 
FFm1~0*1 #the intercept of FFm1 is 0 
FFt1~0*1 #the intercept of FFt1 is 0

FFa1~~phia*FFa1 #variance of FFa1 is phia
FFm1~~phim*FFm1 #variance of FFm1 is phim
FFt1~~phit*FFt1 #variance of FFt1 is phit

FFa1~~phiat*FFt1 #covariance of FFa1 and FFt1 is phiat (yellow curved arrow)
FFa1~~phiam*FFm1 #covariance of FFa1 and FFm1 is phiam (yellow curved arrow)

#FFt1~~phita*FFa1 #covariance of FFt1 and FFa1 is phita (yellow curved arrow) #error: duplicate model element
FFt1~~phitm*FFm1 #covariance of FFt1 and FFm1 is phitm (yellow curved arrow)

#FFm1~~phima*FFa1 #covariance of FFm1 and FFa1 is phima (yellow curved arrow) #error: duplicate model element
#FFm1~~phimt*FFt1 #covariance of FFm1 and FFt1 is phimt (yellow curved arrow) #error: duplicate model element


FFm2~ betam*FFm1+gammam*FFa1 
FFt2~ betat*FFt1+epsilont*FFm2 
FFa2~ betaa*FFa1+zetaa*FFt2                 #what about a1? and t1? answer: cannot predict the 1st time

FFm3~ betam*FFm2+gammam*FFa2
FFt3~ betat*FFt2+epsilont*FFm3
FFa3~ betaa*FFa2+zetaa*FFt3

FFm4~ betam*FFm3+gammam*FFa3 
FFt4~ betat*FFt3+epsilont*FFm4
FFa4~ betaa*FFa3+zetaa*FFt4

FFm5~ betam*FFm4+gammam*FFa4 
FFt5~ betat*FFt4+epsilont*FFm5
FFa5~ betaa*FFa4+zetaa*FFt5

FFm6~ betam*FFm5+gammam*FFa5 
FFt6~ betat*FFt5+epsilont*FFm6
FFa6~ betaa*FFa5+zetaa*FFt6

FFm7~ betam*FFm6+gammam*FFa6 
FFt7~ betat*FFt6+epsilont*FFm7
FFa7~ betaa*FFa6+zetaa*FFt7


FFa2~~Omegaa*FFa2 #variance of FFa2 is Omegaa
FFa3~~Omegaa*FFa3 
FFa4~~Omegaa*FFa4 
FFa5~~Omegaa*FFa5 
FFa6~~Omegaa*FFa6 
FFa7~~Omegaa*FFa7 


FFm2~~Omegam*FFm2 #variance of FFm2 is Omegam
FFm3~~Omegam*FFm3 
FFm4~~Omegam*FFm4 
FFm5~~Omegam*FFm5 
FFm6~~Omegam*FFm6 
FFm7~~Omegam*FFm7 


FFt2~~Omegat*FFt2 #variance of FFt2 is Omegat
FFt3~~Omegat*FFt3 
FFt4~~Omegat*FFt4 
FFt5~~Omegat*FFt5 
FFt6~~Omegat*FFt6 
FFt7~~Omegat*FFt7 



FFa2~~Omegaam*FFm2 #covariance of FFa2 and FFm2 is Omegaam
FFa3~~Omegaam*FFm3 # 24-03-15: Abe and I found out somehow we skipped this
FFa4~~Omegaam*FFm4 # 24-03-15: added these covariances
FFa5~~Omegaam*FFm5
FFa6~~Omegaam*FFm6
FFa7~~Omegaam*FFm7
FFa2~~Omegaat*FFt2
FFa3~~Omegaat*FFt3
FFa4~~Omegaat*FFt4
FFa5~~Omegaat*FFt5
FFa6~~Omegaat*FFt6
FFa7~~Omegaat*FFt7
FFt2~~Omegatm*FFm2
FFt3~~Omegatm*FFm3
FFt4~~Omegatm*FFm4
FFt5~~Omegatm*FFm5
FFt6~~Omegatm*FFm6
FFt7~~Omegatm*FFm7

FFm1 =~ 1*m1 #define the latent variable f*m1 (from fm) #here Abe suggests to also freely estimate insteading of fixing the parameters to 1
FFm2 =~ 1*m2 
FFm3 =~ 1*m3 
FFm4 =~ 1*m4 
FFm5 =~ 1*m5 
FFm6 =~ 1*m6 
FFm7 =~ 1*m7 

FFa1 =~ 1*a1 #define the latent variable a*1 (from fa)
FFa2 =~ 1*a2 
FFa3 =~ 1*a3 
FFa4 =~ 1*a4 
FFa5 =~ 1*a5 
FFa6 =~ 1*a6 
FFa7 =~ 1*a7 

FFt1 =~ 1*t1 #define the latent variable t*1 (from ft) currently not in the graph
FFt2 =~ 1*t2 
FFt3 =~ 1*t3 
FFt4 =~ 1*t4 
FFt5 =~ 1*t5 
FFt6 =~ 1*t6 
FFt7 =~ 1*t7 

'