Dear all

I have a list of coefficient and standard error averages (following my previous post 1576600-table-after-tuples-forval to be more exact). I've run a new model taking one of the variables out and I would like to see how my (average) coefficients changed. I was hoping to use command -suest- to see if my column b in model 1 is significantly different from my column b on model 2.

Model 1:
Code:
set seed 12345

. set obs 100
number of observations (_N) was 7, now 100

. local dvar y

. gen `dvar' = runiformint(0, 1)

. local always_vars x1 x2 x3

. foreach a of local always_vars {
  2.     gen `a' = rnormal()
  3. }

. local tlist w1 w2 w3 w4

. foreach t of local tlist {
  2.     gen `t' = rnormal()
  3. }

. 
. local nvars: word count `tlist'

. 
. local ntuples = 0

. 
. forvalues i = 1/`nvars' {
  2.     forvalues j = `=`i'+1'/`nvars' {
  3.         local v1: word `i' of `tlist'
  4.         local v2: word `j' of `tlist'
  5.         local ++ntuples
  6.         local tuple`ntuples' `v1' `v2'
  7.     }
  8. }

. 
. 
. 
.
. tempfile results

. 
. capture postutil clear

. 
. local results_vars

. foreach v of varlist `always_vars' `tlist' {
  2.     local results_vars `results_vars' b_`v' se_`v'
  3. }

. 
. local initial

. foreach v of varlist `always_vars' {
  2.     local initial `initial' (_b[`v']) (_se[`v'])
  3. }

. 
. postfile handle `results_vars' using `results'

. 
.     
. forvalues i = 1/`ntuples' {
  2.      display `"`tuple`i''"'
  3.      quietly regress `dvar' `always_vars' `tuple`i'', vce(robust)
  4.      if c(rc) == 0 {
  5.          local topost `initial'
  6.          foreach v of local tlist {
  7.              if strpos("`e(cmdline)'", "`v'") {
  8.                  local topost `topost' (_b[`v']) (_se[`v'])
  9.              }
 10.              else {
 11.                  local topost `topost' (.) (.)
 12.              }
 13.          }
 14.          post handle `topost'
 15.      }
 16.      else if inlist(c(rc), 2000, 2001) { // NO, OR TOO FEW OBSERVATIONS
 17.          continue
 18.      }
 19.      else { //  REGRESSION FAILED DUE UNEXPECTED PROBLEM: ABORT
 20.          display as error "Unexpected error with `tuple`i''"
 21.          exit(c(rc))
 22.      }
 23.              
. }
w1 w2
w1 w3
w1 w4
w2 w3
w2 w4
w3 w4

. 
. postclose handle

. 

. use `results', clear

. 
. 
. local critical_value_5 = invnormal(0.975)

. 
. foreach v in `always_vars' `tlist' {
  2.      gen byte sig05_`v' = abs(b_`v'/se_`v') > `critical_value_5' ///
>         if !missing(b_`v', se_`v')
  3.     order sig05_`v', after(se_`v')
  4. }
(3 missing values generated)
(3 missing values generated)
(3 missing values generated)
(3 missing values generated)

. 
.
. collapse (mean) _all

. 
. gen one = 1

. 
. reshape long b_ se_ sig05_, i(one) j(varname) string
(note: j = w1 w2 w3 w4 x1 x2 x3)

Data                               wide   ->   long
-----------------------------------------------------------------------------
Number of obs.                        1   ->       7
Number of variables                  22   ->       5
j variable (7 values)                     ->   varname
xij variables:
                     b_w1 b_w2 ... b_x3   ->   b_
                  se_w1 se_w2 ... se_x3   ->   se_
         sig05_w1 sig05_w2 ... sig05_x3   ->   sig05_
-----------------------------------------------------------------------------

. 
. rename *_ *

. 
. drop one

. 
. list, noobs clean

    varname           b         se   sig05  
         w1   -.0421191    .045018       0  
         w2   -.0723879   .0540686       0  
         w3    .0286007   .0493419       0  
         w4    .0096308   .0488577       0  
         x1   -.0032636   .0469172       0  
         x2    .0817245   .0491854       0  
         x3   -.0240363   .0476184       0  

est store m1


Model 2 (The same code as model 1 but without w4 on the tlist):
Code:
list, noobs clean

    varname           b         se   sig05  
         w2   -.0424746   .0449553       0  
         w3   -.0717693    .054142       0  
         w4    .0291465   .0490506       0  
         x1   -.0063258   .0469255       0  
         x2       .0796   .0493231       0  
         x3    -.025599    .047796       0

est store m2
However when I try to use the -suest- command, I get the following error:

Code:
suest m1 m2
estimation sample of the model saved under m1 could not be restored
Am I doing something wrong? Should I use another command to check for the significance of differences between models?

Thank you in advance