我正在使用 sem 包来构建 SEM 模型。该模型可以运行，但是会出现如下警告消息：
Warning message:
In eval(expr, envir, enclos) :
Could not compute QR decomposition of Hessian.
Optimization probably did not converge.
如果我运行 summary(cfa1.test)，警告错误消息是这样的：
Error in summary.objectiveML(cfa1.test) :
coefficient covariances cannot be computed
In addition: Warning message:
In vcov.sem(object, robust = robust, analytic = analytic.se) :
singular Hessian: model is probably underidentified.
我的模型是关于市场导向（MO）、政府激励（GvIncent）、创业导向（EO）和企业绩效（FirmP）之间的关系。这些是四个二阶构造。在它们中的每一个下都有许多构造，例如，响应性 (Resp)、主动性 (Proa)、经济绩效 (EconP)、政府财政激励 (GvFin)，在这些子构造下，有观察到的变量，例如 Resp3、 4、5 和 EconP（1-12）。
以下是我的模型的代码：
model.1<-specifyModel()
GvInfo->GvInfo1, NA, 1 #all the lamda observed to latent (56 lines)
GvInfo->GvInfo2, lam112, NA
GvInfo->GvInfo3, lam113, NA
GvInfo->GvInfo4, lam114, NA
GvInfo->GvInfo5, lam115, NA
GvInfo->GvInfo6, lam116, NA
GvInfo->GvInfo7, lam117, NA
GvFin->GvFin1, NA, 1
GvFin->GvFin2, lam122, NA
GvFin->GvFin3, lam123, NA
GvFin->GvFin4, lam124, NA
GvFin->GvFin5, lam125, NA
GvFin->GvFin6, lam126, NA
GvFin->GvFin7, lam127, NA
GvPat->GvPat1, NA, 1
GvPat->GvPat2, lam132, NA
GvPat->GvPat3, lam133, NA
GvPat->GvPat4, lam134, NA
GvPat->GvPat5, lam135, NA
EconP->EconP1, NA, 1
EconP->EconP2, lam402, NA
EconP->EconP3, lam403, NA
EconP->EconP4, lam404, NA
EconP->EconP5, lam405, NA
EconP->EconP6, lam406, NA
EconP->EconP7, lam407, NA
EconP->EconP8, lam408, NA
EconP->EconP9, lam409, NA
EconP->EconP10, lam410, NA
EconP->EconP11, lam411, NA
EconP->EconP12, lam412, NA
SocP->SocP1, NA, 1
SocP->SocP2, lam422, NA
InfoG->InfoG1, NA, 1
InfoG->InfoG2, lam312, NA
InfoG->InfoG3, lam313, NA
InfoG->InfoG4, lam314, NA
InfoG->InfoG5, lam315, NA
InfoG->InfoG6, lam316, NA
InfoD->InfoD1, NA, 1
InfoD->InfoD2, lam322, NA
InfoD->InfoD3, lam323, NA
InfoD->InfoD4, lam324, NA
InfoD->InfoD5, lam325, NA
Resp->Resp3, NA, 1
Resp->Resp4, lam334, NA
Resp->Resp5, lam335, NA
Innov->Innov1, NA, 1
Innov->Innov2, lam212, NA
Innov->Innov3, lam213, NA
Proa->Proa1, NA, 1
Proa->Proa2, lam222, NA
Proa->Proa3, lam223, NA
RiskT->RiskT1, NA, 1
RiskT->RiskT2, lam232, NA
RiskT->RiskT3, lam233, NA #all the lamda observed to latent
GvIncent->GvInfo, NA, 1 #sub construct to construct (11 lines)
GvIncent->GvFin, beta12, NA
GvIncent->GvPat, beta13, NA
FirmP->EnoP, NA, 1
FirmP->SocP, beta42, NA
EO->Innov, NA, 1
EO->Proa, beta22, NA
EO->RiskT, beta23, NA
MO->InfoG, NA, 1
MO->InfoD, beta32, NA
MO->Resp, beta33, NA #sub construct to construct
EO->FirmP, beta24, NA #path from end to end constructs (5 lines)
MO->FirmP, beta34, NA #path from end to end constructs
GvIncent->EO, gam12, NA #exd to end construct
GvIncent->FirmP, gam14, NA #exd to end construct
EO<->MO, psi23, NA #correlation btw end
GvIncent<->GvIncent, phi11, NA #error temrs for exds (4 lines)
EO<->EO, psi22, NA #error temrs for ends
MO<->MO, psi33, NA
FirmP<->FirmP, psi44, NA #error temrs for ends
GvInfo<->GvInfo, psi1111, NA #error temrs for ends subconstructs (11 lines)
GvFin<->GvFin, psi1212, NA
GvPat<->GvPat, psi1313, NA
Innov<->Innov, psi2121, NA
Proa<->Proa, psi2222, NA
RiskT<->RiskT, psi2323, NA
InfoG<->InfoG, psi3131, NA
InfoD<->InfoD, psi3232, NA
Resp<->Resp, psi3333, NA #error terms for ends subconstructs
EconP<->EconP, psi4141, NA
SocP<->SocP, psi4242, NA
GvInfo1<->GvInfo1, the111, NA #errors of the factor indicators of ends
GvInfo2<->GvInfo2, the112, NA
GvInfo3<->GvInfo3, the113, NA
GvInfo4<->GvInfo4, the114, NA
GvInfo5<->GvInfo5, the115, NA
GvInfo6<->GvInfo6, the116, NA
GvInfo7<->GvInfo7, the117, NA
GvFin1<->GvFin1, the121, NA
GvFin2<->GvFin2, the122, NA
GvFin3<->GvFin3, the123, NA
GvFin4<->GvFin4, the124, NA
GvFin5<->GvFin5, the125, NA
GvFin6<->GvFin6, the126, NA
GvFin7<->GvFin7, the127, NA
GvPat1<->GvPat1, the131, NA
GvPat2<->GvPat2, the132, NA
GvPat3<->GvPat3, the133, NA
GvPat4<->GvPat4, the134, NA
GvPat5<->GvPat5, the135, NA
EconP1<->EconP1, the401, NA
EconP2<->EconP2, the402, NA
EconP3<->EconP3, the403, NA
EconP4<->EconP4, the404, NA
EconP5<->EconP5, the405, NA
EconP6<->EconP6, the406, NA
EconP7<->EconP7, the407, NA
EconP8<->EconP8, the408, NA
EconP9<->EconP9, the409, NA
EconP10<->EconP10, the410, NA
EconP11<->EconP11, the411, NA
EconP12<->EconP12, the412, NA
SocP1<->SocP1, the421, NA
SocP2<->SocP2, the422, NA
InfoG1<->InfoG1, the311, NA
InfoG2<->InfoG2, the312, NA
InfoG3<->InfoG3, the313, NA
InfoG4<->InfoG4, the314, NA
InfoG5<->InfoG5, the315, NA
InfoG6<->InfoG6, the316, NA
InfoD1<->InfoD1, the321, NA
InfoD2<->InfoD2, the322, NA
InfoD3<->InfoD3, the323, NA
InfoD4<->InfoD4, the324, NA
InfoD5<->InfoD5, the325, NA
Resp3<->Resp3, the333, NA
Resp4<->Resp4, the334, NA
Resp5<->Resp5, the335, NA
Innov1<->Innov1, the211, NA
Innov2<->Innov2, the212, NA
Innov3<->Innov3, the213, NA
Proa1<->Proa1, the221, NA
Proa2<->Proa2, the222, NA
Proa3<->Proa3, the223, NA
RiskT1<->RiskT1, the231, NA
RiskT2<->RiskT2, the232, NA
RiskT3<->RiskT3, the233, NA #errors of the factor indicators of ends
#model1 test
cfa1.test<-sem(model.1,cormat.model.1, N=534)
summary(cfa1.test)
如果有人能让我知道错误和警告消息指示哪些问题以及我应该如何处理它们，那就太好了。任何帮助将不胜感激。