Different Outputs for the same programs

[Tung Luong]

Hello together!

I have a STATA program that I got from a research paper and I modified it a little bit. I run the program and it returned the outcome I need. However, when I copied it to another folder and run it again, it just produced a totally different outcome. Do you have any suggestions to fix it ?

To make it clear: I copied both the program file and the data file to the new folder, so basically I have EXACTLY the same programs and datasets in the 2 folder. I also changed the directory in the new copied file to the new folder already. I am using STATA 15.

Here you can find the program and dataset, along with an image of the output I want.

I'm totally new to STATA so my question my be vague ... I hope I can make it as clear as possible if you have any questions.

Thanks in advance!

[Carlo Lazzaro]

Tung:
welcome to this forum.
As per FAQ, the best way to allow other to replicate your issue(s), is to:
-shared an example/excerpt of your dataset via -dataex- (as you know, downloading files is not risk-free);
-using CODE delimiters to post what you typed and what Stata gave you back in both instances.
Thanks.

[Tung Luong]

Thanks Carlo for pointing it out!

Here is what I want, and also what I got from the original program:

Here is what I got after copying the program and data files to another folder, which is not what I expected because I'm using the same program and dataset:

Hereunder are my program and a part of my dataset:

clear all

*** Replicate Romer and Romer (2017) 
*** Folder & data
cd "E:/Goethe/GOETHE courses/Seminar/Banking and Macro/Replication - Topic 9/Figure 2 - GDP"
use data

*** Country and time variables
gen time = _n
tsset time
xtset id date
 
 *** Dependent variable
gen y = ln(gdp)
local lhs "y"   /* LeftHandSide variables: variables that IRFs are created for */
local H = 11            /* Impulse response horizon */
local lag = 4       /* Lag length */  
 
 *** Crisis variable
gen x = crisis
gen z = ln(gdp)

foreach v in `lhs'  {
quietly:    tsrevar F(0/`H').`v'
quietly:    rename (`r(varlist)') `v'_#, addnumber
quietly:    gen `v' = `v'_1
}

* 3.7 EA, 7 US
foreach v in `lhs'  {
forvalues i=1/`H' {
xtreg `v'_`i' L(0/`lag').x L(1/`lag').y i.date i.id, fe robust 
quietly: est sto irf_`v'_`i'
* Rescale by 7, i.e. effect of a moderate financial crisis-minus
quietly: generate beta_`v'_`i' = _b[x]*7
quietly: generate se_`v'_`i' = _se[x]*7
}
}
 
*preserve
keep beta* se*
drop if _n > 1
gen i=_n
reshape long beta_y_ se_y_, i(i) j(time)
gen t = time -1

foreach v in `lhs'  {
gen ub_`v' = beta_`v'_ + 1.64*se_`v'_ /* 90% CI: beta +/- 1.64*SE */
gen lb_`v' = beta_`v'_ - 1.64*se_`v'_ 
twoway (line beta_`v'_ t ) (line ub_`v' t ) (line lb_`v' t), name(symplot_`v')  legend(off) ytitle(`v')
}

* Example generated by -dataex-. To install: ssc install dataex
clear
input int date str14 country byte id float gdp byte crisis
 1 "Australia" 1  215022.7 .
 2 "Australia" 1  221020.9 .
 3 "Australia" 1 227456.03 .
 4 "Australia" 1  238985.5 .
 5 "Australia" 1  242319.8 .
 6 "Australia" 1 251883.36 .
 7 "Australia" 1 261735.63 .
 8 "Australia" 1 265908.53 .
 9 "Australia" 1 266415.72 .
10 "Australia" 1    274636 0
11 "Australia" 1  277648.9 0
12 "Australia" 1 278936.94 0
13 "Australia" 1 286702.78 0
14 "Australia" 1  296848.8 0
15 "Australia" 1  290835.5 0
16 "Australia" 1  294636.8 0
17 "Australia" 1 304903.38 0
18 "Australia" 1 297140.06 0
19 "Australia" 1  311762.8 0
20 "Australia" 1  316578.5 0
21 "Australia" 1  319310.2 0
22 "Australia" 1  316857.2 0
23 "Australia" 1  322089.6 0
24 "Australia" 1  329340.7 0
25 "Australia" 1  333119.4 0
26 "Australia" 1  342733.2 0
27 "Australia" 1  345098.3 0
28 "Australia" 1  353017.3 0
29 "Australia" 1  359708.5 0
30 "Australia" 1  365543.6 0
31 "Australia" 1    365988 0
32 "Australia" 1  357717.5 0
33 "Australia" 1    353286 0
34 "Australia" 1  369385.1 0
35 "Australia" 1 382245.25 0
36 "Australia" 1  389197.6 0
37 "Australia" 1    403278 0
38 "Australia" 1  408020.8 0
39 "Australia" 1  409823.6 0
40 "Australia" 1  418003.7 0
41 "Australia" 1  428827.6 0
42 "Australia" 1  445082.4 0
43 "Australia" 1  448268.6 0
44 "Australia" 1  457975.2 0
45 "Australia" 1  472645.7 0
46 "Australia" 1  475018.3 0
47 "Australia" 1    479593 0
48 "Australia" 1  480137.8 0
49 "Australia" 1 472552.75 0
50 "Australia" 1  475118.8 0
51 "Australia" 1  482585.8 0
52 "Australia" 1 496786.75 0
53 "Australia" 1    503596 0
54 "Australia" 1  513588.8 0
55 "Australia" 1  528575.7 0
56 "Australia" 1 536241.06 0
57 "Australia" 1 541486.06 0
58 "Australia" 1    553583 0
59 "Australia" 1  565288.2 0
60 "Australia" 1  576162.4 0
61 "Australia" 1 595942.25 0
62 "Australia" 1    605320 0
63 "Australia" 1  616236.8 0
64 "Australia" 1  635587.4 0
65 "Australia" 1  643985.9 0
66 "Australia" 1  658289.8 0
67 "Australia" 1  670424.4 0
68 "Australia" 1  668538.8 0
69 "Australia" 1  681529.5 0
70 "Australia" 1  696473.6 0
71 "Australia" 1  713142.7 0
72 "Australia" 1  721937.9 0
73 "Australia" 1  726673.2 0
74 "Australia" 1  750849.4 0
75 "Australia" 1  760423.1 0
76 "Australia" 1  772823.8 0
77 "Australia" 1  783484.5 0
78 "Australia" 1  799169.3 0
79 "Australia" 1  802247.5 0
80 "Australia" 1  824043.6 0
81 "Australia" 1  842871.8 0
82 "Australia" 1  854998.9 0
83 "Australia" 1  867274.1 2
84 "Australia" 1  867477.4 4
85 "Australia" 1    881829 1
86 "Australia" 1  890988.3 0
87 "Australia" 1  900192.8 0
88 "Australia" 1  914895.9 0
89 "Australia" 1  922691.9 0
90 "Australia" 1  943332.9 0
91 "Australia" 1  958229.4 0
92 "Australia" 1  968897.6 0
 1 "Austria"   2 109520.88 0
 2 "Austria"   2 110551.05 0
 3 "Austria"   2 113272.77 0
 4 "Austria"   2 116357.67 0
 5 "Austria"   2  119713.9 0
 6 "Austria"   2 123842.02 0
 7 "Austria"   2 129449.08 0
 8 "Austria"   2 133593.97 0
end