access to the full set of parameters of one single equation

[Alvaro Gutierrez]

Dear Statalist,

I have a simple question regarding how to get access to the full set of parameters that belong to one single equation. To exemplify, below there is a code that fits a two-equation model using -ml-. What I want to obtain is a row vector containing all of the parameters from the mu equation (3 regressors). Is there any direct way to obtain this using the name of the equation?

I know that we can access to specific parameters using the following syntax: _b[mu:weight]. Unfortunately, I couldn't find a way to get access to the full set of parameters associated with the mu equation.

*** mynormal1_lf.ado ***

program mynormal1_lf
version 11
args lnfj mu sigma
quietly replace `lnfj' = ln(normalden($ML_y1,`mu',`sigma'))
end

*** runmynormal.do ***

. sysuse auto, clear 
(1978 Automobile Data)

. ml model lf mynormal1_lf (mu: mpg = weight displacement) (sigma:)

. ml maximize

initial:       log likelihood =     -<inf>  (could not be evaluated)
feasible:      log likelihood = -10383.274
rescale:       log likelihood = -292.89564
rescale eq:    log likelihood = -238.45986
Iteration 0:   log likelihood = -238.45986  (not concave)
Iteration 1:   log likelihood = -225.48871  (not concave)
Iteration 2:   log likelihood = -215.66715  
Iteration 3:   log likelihood = -199.18204  
Iteration 4:   log likelihood = -195.29955  
Iteration 5:   log likelihood =    -195.24  
Iteration 6:   log likelihood =  -195.2398  
Iteration 7:   log likelihood =  -195.2398  

                                                Number of obs     =         74
                                                Wald chi2(2)      =     139.21
Log likelihood =  -195.2398                     Prob > chi2       =     0.0000

------------------------------------------------------------------------------
         mpg |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
mu           |
      weight |  -.0065671   .0011424    -5.75   0.000    -.0088061   -.0043281
displacement |   .0052808   .0096674     0.55   0.585    -.0136671    .0242286
       _cons |   40.08452   1.978738    20.26   0.000     36.20627    43.96278
-------------+----------------------------------------------------------------
sigma        |
       _cons |   3.385282   .2782684    12.17   0.000     2.839886    3.930678
------------------------------------------------------------------------------

. di _b[mu:weight]
-.00656711

Thanks in advance for your time.
Álvaro

---

Stata 16.1 MP
Win10/Linux Mint 19.1
https://alvarogutyerrez.github.io/

[Andrew Musau]

Estimation results are stored in e. So you just need to extract the desired vector (coefficients) or matrix (variances).

cap drop mynormal1_lf
program mynormal1_lf
version 11
args lnfj mu sigma
quietly replace `lnfj' = ln(normalden($ML_y1,`mu',`sigma'))
end
sysuse auto, clear
ml model lf mynormal1_lf (mu: mpg = weight displacement) (sigma:)
ml maximize
mat mub = e(b)[1, "mu:"]
mat muV= e(V)["mu:", "mu:"]

Res.:

. mat l mub

mu[1,3]
              mu:           mu:           mu:
          weight  displacement         _cons
y1    -.00656711     .00528078     40.084522



. mat list muV

symmetric muV[3,3]
                           mu:           mu:           mu:
                       weight  displacement         _cons
      mu:weight     1.305e-06
mu:displacement    -9.883e-06     .00009346
       mu:_cons    -.00199045     .01140179      3.915406

[Alvaro Gutierrez]

Thank you a lot Andrew Musau. I totally forgot that the e(b) matrix is labeled with equation names!

Best regards,

Álvaro

---

Stata 16.1 MP
Win10/Linux Mint 19.1
https://alvarogutyerrez.github.io/