Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# st: normality test using the over identifying moment conditions

 From Usman Gilani To "statalist@hsphsun2.harvard.edu" Subject st: normality test using the over identifying moment conditions Date Thu, 31 Jan 2013 12:21:55 +0000

 Hi,I'm trying to interpret the following results, with respect to "normality test using the over identifying moment conditions"where returns have normal distributionwith parameter mu,sdand i have 4 moment conditions >E[r-mu/sd]=0>E[(r-mu)^2/sd-1]=0>E[(r-mu)^3/sd^3]=0>E[(r-mu)^4/sd^4-3]=0output..gel(g = g, x = returns, tet0 = c(f3\$estimate[1], f3\$estimate[2]))Type of GEL:  EL Coefficients:              Estimate  Std. Error     t value   Pr(>|t|)mean  -0.01168   0.05614    -0.20805   0.83519sd         1.77591   0.03965    44.79218   0.00000Lambdas:                        Estimate   Std. Error  t value    Pr(>|t|) Lambda[1]   -0.09743    0.03912    -2.49028    0.01276Lambda[2]    0.65728    0.02443    26.90505    0.00000Lambda[3]    0.03247    0.01304     2.48961    0.01279Lambda[4]   -0.10954    0.00407   -26.90423    0.00000 Over-identifying restrictions tests: degrees of freedom is 2                      statistics     p-value    LR test   2.3341e+02   2.0730e-51LM test   7.2417e+02   5.5954e-158J test      7.2417e+02    5.5954e-158Convergence code for the coefficients:  0 Convergence code for the lambdas:  0 does the J-test p-value rejecting the null E[g(theta,x)]=0, and which moment condition is true under normality