Dropping 0s in Matrices

[Jared Greathouse]

Note that I use synth from SSC. Consider the following

cls
u "http://fmwww.bc.edu/repec/bocode/s/scul_basque.dta", clear

qui xtset
local lbl: value label `r(panelvar)'

loc unit ="Basque Country (Pais Vasco)":`lbl'


loc int_time = 1975

qui xtset
cls

loc x 1961(1)1969

loc y 1960(1)1969

loc z 1964(1)1969



tempfile scmbasque scmbasque2
cls
synth gdpcap /// classic SCM
    gdpcap(`y') ///
    pop(1969) ///
        sec_agriculture(`x') ///
        sec_energy(`x') ///
        sec_industry(`x') ///
        sec_construction(`x') ///
        sec_svc_venta(`x') ///
        sec_svc_nonventa(`x') ///
    school_illit(`z') ///
    school_prim(`z') ///
    school_med(`z') ///
    school_high(`z') ///
    invest(`z'), ///
    trunit(`unit') ///
    trperiod(`int_time') ///
    fig nested allopt keep(`scmbasque')


mat li e(V_matrix)

The resulting matrix

symmetric e(V_matrix)[13,13]
              gdpcap~1969)     pop(1969)  sec_ag~1969)  sec_en~1969)  sec_in~1969)  sec_co~1969)  sec_sv~1969)
gdpcap~1969)     .28494051
   pop(1969)             0     .08034821
sec_ag~1969)             0             0     .01250346
sec_en~1969)             0             0             0     .15639678
sec_in~1969)             0             0             0             0     .15177799
sec_co~1969)             0             0             0             0             0     .00277837
sec_sv~1969)             0             0             0             0             0             0     .00470292
sec_sv~1969)             0             0             0             0             0             0             0
school~1969)             0             0             0             0             0             0             0
school~1969)             0             0             0             0             0             0             0
school~1969)             0             0             0             0             0             0             0
school~1969)             0             0             0             0             0             0             0
invest~1969)             0             0             0             0             0             0             0

              sec_sv~1969)  school~1969)  school~1969)  school~1969)  school~1969)  invest~1969)
sec_sv~1969)     .00466699
school~1969)             0     .01746123
school~1969)             0             0     .00170478
school~1969)             0             0             0     .20263447
school~1969)             0             0             0             0      .0674579
invest~1969)             0             0             0             0             0     .01262639

is a 13 by 13 revulsion that looks ugly to me. Shame on the coders for making it be like that- but enough ranting. My question is, how would I delete the 0s from this matrix such that we have something like

                       Predictor | V Weight
-----------------------------+------------
                   gdpcap~1969) |           .28494051
                      pop(1969) |           .08034821

and so on and so forth for each predictor. A colleague wanted to know how to present the matrix in his paper, but, I wanted to show him a way that gives an informative, clear answer, and the present way the matrix looks is, to me, cumbersome. How might I tackle this?

[Andrew Musau]

The resulting matrix is a 13 by 13 revulsion that looks ugly to me. Shame on the coders for making it be like that

To be fair to them, they are displaying the variance matrix which has a very specific structure. You have variances as the diagonal elements and the off-diagonal elements are covariances.

mat wanted= e(V_matrix)[1..., 1]
forval i= 2/ `=colsof(e(V_matrix))'{
    mat wanted= wanted+e(V_matrix)[1..., `i']
}
mat colname wanted= variance

Res.:

. mat l wanted

wanted[13,1]
               variance
gdpcap~1969)  .28494051
   pop(1969)  .08034821
sec_ag~1969)  .01250346
sec_en~1969)  .15639678
sec_in~1969)  .15177799
sec_co~1969)  .00277837
sec_sv~1969)  .00470292
sec_sv~1969)  .00466699
school~1969)  .01746123
school~1969)  .00170478
school~1969)  .20263447
school~1969)   .0674579
invest~1969)  .01262639