lisrel 为什么画图生成语句后，再运行验证性因子分析时说模型不拟合啊？
[i] 本帖最后由 ytstyan 于 2011-12-25 19:36 编辑 [/i]每次运行那个e小人跑步的L时，都说the model does not converge…………
有截图……是因为数据不对么……
老师给的课题，我设置了13个观察指标与4个潜变量。自己模拟了60组数据
是不是因为某个潜变量的观察指标对应的太多了？ 或者自己的数据太少了，或者不准确？？？
PHI is not positive definite!!!!!!!这个错误是为啥啊！！！！！！
W_A_R_N_I_N_G: PHI is not positive definite TI
Parameter Specifications
LAMBDA-X
1 2 3 4
-------- -------- -------- --------
FDSNA 1 0 0 0
FDSAS 2 0 0 0
FDSWD 3 0 0 0
FGSAS 4 0 0 0
FFSNA 5 0 0 0
FFSAS 6 0 0 0
FFSWD 7 0 0 0
FISNA 8 0 0 0
CBSWD 0 9 0 0
CCSNA 0 10 0 0
CMSAS 0 11 0 0
FIDSNA 0 0 12 0
FTDSTOLS 0 0 13 0
FEDSNA 0 0 14 0
FRDSMOD 0 0 15 0
LDSFIV 0 0 0 16
LMSNA 0 0 0 17
LMSAS 0 0 0 18
LMSWD 0 0 0 19
LTSNA 0 0 0 20
LTSAS 0 0 0 21
LTSWD 0 0 0 22
LRSNA 0 0 0 23
LRSAS 0 0 0 24
LRSWD 0 0 0 25
PHI
1 2 3 4
-------- -------- -------- --------
1 0
2 26 0
3 27 28 0
4 29 30 31 0
THETA-DELTA
FDSNA FDSAS FDSWD FGSAS FFSNA FFSAS
-------- -------- -------- -------- -------- --------
32 33 34 35 36 37
THETA-DELTA
FFSWD FISNA CBSWD CCSNA CMSAS FIDSNA
-------- -------- -------- -------- -------- --------
38 39 40 41 42 43
THETA-DELTA
FTDSTOLS FEDSNA FRDSMOD LDSFIV LMSNA LMSAS
-------- -------- -------- -------- -------- --------
44 45 46 47 48 49
THETA-DELTA
LMSWD LTSNA LTSAS LTSWD LRSNA LRSAS
-------- -------- -------- -------- -------- --------
50 51 52 53 54 55
THETA-DELTA
LRSWD
--------
56
W_A_R_N_I_N_G: PHI is not positive definite
W_A_R_N_I_N_G: THETA-DELTA is not positive definite
TI
W_A_R_N_I_N_G: The solution was found non-admissible after 50 iterations.
The following solution is preliminary and is provided only
for the purpose of tracing the source of the problem.
Setting AD> 50 or AD=OFF may solve the problem
LISREL Estimates(Intermediate Solution)
LAMBDA-X
1 2 3 4
-------- -------- -------- --------
FDSNA 1.81 - - - - - -
FDSAS 0.04 - - - - - -
FDSWD 0.70 - - - - - -
FGSAS 0.14 - - - - - -
FFSNA -0.49 - - - - - -
FFSAS -0.17 - - - - - -
FFSWD -0.29 - - - - - -
FISNA -0.23 - - - - - -
CBSWD - - 1.85 - - - -
CCSNA - - 0.28 - - - -
CMSAS - - -0.65 - - - -
FIDSNA - - - - -0.58 - -
FTDSTOLS - - - - -0.08 - -
FEDSNA - - - - -0.13 - -
FRDSMOD - - - - -0.53 - -
LDSFIV - - - - - - -0.21
LMSNA - - - - - - 0.20
LMSAS - - - - - - 0.23
LMSWD - - - - - - 0.04
LTSNA - - - - - - 0.24
LTSAS - - - - - - 0.40
LTSWD - - - - - - -0.35
LRSNA - - - - - - -0.06
LRSAS - - - - - - -0.07
LRSWD - - - - - - -0.70
PHI
1 2 3 4
-------- -------- -------- --------
1 1.00
2 -0.16 1.00
3 -0.31 -0.73 1.00
4 -0.30 -0.86 3.23 1.00
W_A_R_N_I_N_G: PHI is not positive definite
THETA-DELTA
FDSNA FDSAS FDSWD FGSAS FFSNA FFSAS
-------- -------- -------- -------- -------- --------
-1.63 1.58 1.69 1.79 2.78 2.22
THETA-DELTA
FFSWD FISNA CBSWD CCSNA CMSAS FIDSNA
-------- -------- -------- -------- -------- --------
2.09 1.78 -0.79 2.01 1.94 2.55
THETA-DELTA
FTDSTOLS FEDSNA FRDSMOD LDSFIV LMSNA LMSAS
-------- -------- -------- -------- -------- --------
2.46 2.13 2.38 2.45 2.29 1.81
THETA-DELTA
LMSWD LTSNA LTSAS LTSWD LRSNA LRSAS
-------- -------- -------- -------- -------- --------
1.96 2.95 2.25 2.54 2.59 2.15
THETA-DELTA
LRSWD
--------
1.34
Goodness of Fit Statistics
Degrees of Freedom = 269
Minimum Fit Function Chi-Square = 643.57 (P = 0.0)
Normal Theory Weighted Least Squares Chi-Square = 303.21 (P = 0.074)
Estimated Non-centrality Parameter (NCP) = 34.21
90 Percent Confidence Interval for NCP = (0.0 ; 81.13)
90 Percent Confidence Interval for F0 = (0.0 ; 3.38)
Root Mean Square Error of Approximation (RMSEA) = 0.073
90 Percent Confidence Interval for RMSEA = (0.0 ; 0.11)
P-Value for Test of Close Fit (RMSEA < 0.05) = 0.23
Expected Cross-Validation Index (ECVI) = 17.30
90 Percent Confidence Interval for ECVI = (15.88 ; 19.26)
ECVI for Saturated Model = 27.08
ECVI for Independence Model = 17.83
Chi-Square for Independence Model with 300 Degrees of Freedom = 377.84
Independence AIC = 427.84
Model AIC = 415.21
Saturated AIC = 650.00
Independence CAIC = 483.31
Model CAIC = 539.47
Saturated CAIC = 1371.13
Normed Fit Index (NFI) = -0.70
Non-Normed Fit Index (NNFI) = -4.37
Parsimony Normed Fit Index (PNFI) = -0.63
Comparative Fit Index (CFI) = 0.0
Incremental Fit Index (IFI) = -2.44
Relative Fit Index (RFI) = -0.90
Critical N (CN) = 13.15
Root Mean Square Residual (RMR) = 0.45
Standardized RMR = 0.20
Goodness of Fit Index (GFI) = 0.50
Adjusted Goodness of Fit Index (AGFI) = 0.40
Parsimony Goodness of Fit Index (PGFI) = 0.42
Time used: 0.062 Seconds
这是OUT里写的~ W_A_R_N_I_N_G: PHI is not positive definite
W_A_R_N_I_N_G: THETA-DELTA is not positive definite
TI
W_A_R_N_I_N_G: The solution was found non-admissible after 50 iterations.
The following solution is preliminary and is provided only
for the purpose of tracing the source of the problem.
Setting AD> 50 or AD=OFF may solve the problem
有三个警告
页:
[1]