Closed gcasamat closed 12 months ago
I experienced the same error when using jackknife as inference option.
Hello you two, I just solved the problem. Please reinstall from github and let me know if this works for both of you.
Thanks for your prompt response. I am still in the process of running the program using jackknife inference, the problem is that it takes forever to generate the outcome even when I limit my sample.... At least.... ten min in, no error yet. A side question, is there any rough estimates of time running sdid using jackknife inference methods? For example, if SDID estimates without inference takes around 7 min (I have a fairly large dataset, particularly the control group), how much more time should i expect if I would like to use jackknife method? My understanding with Jackknife is that, we use the same weight generated by SDID, but then get the double difference using leave-one-out sample for the entire sample. So the bulk of the time spent is to repeat the OLS process N times? This can be long when N is huuuuuuge. Do you have any suggestion on the optimal sample size that balances the tradeoff between precision and computation burden?
Hi, I confirm that it works fine now with the jackknife method. I however encountered an error message while using the placebo method. Here is the full traceback:
---------------------------------------------------------------------------
SystemError Traceback (most recent call last)
/var/folders/1y/v8whblks715474p0yvpj5s9r0000gn/T/ipykernel_46194/1682810350.py in <module>
----> 1 get_ipython().run_cell_magic('stata', '', "sdid `outcome_var' cod_insee year treated, vce(`se_method') graph g2_opt(scheme(plotplainblind)) seed(123) reps(50)\n*sdid `outcome_var' cod_insee year treated, covariates(`control_varlist', projected) vce(`se_method') graph g2_opt(scheme(plotplainblind)) seed(123) reps(25)\n*sdid `outcome_var' cod_insee year treated, covariates(`control_varlist') vce(bootstrap) graph g2_opt(scheme(plotplainblind)) seed(123) reps(50)\n")
~/opt/anaconda3/envs/py310/lib/python3.10/site-packages/IPython/core/interactiveshell.py in run_cell_magic(self, magic_name, line, cell)
2417 with self.builtin_trap:
2418 args = (magic_arg_s, cell)
-> 2419 result = fn(*args, **kwargs)
2420 return result
2421
~/opt/anaconda3/envs/py310/lib/python3.10/site-packages/decorator.py in fun(*args, **kw)
230 if not kwsyntax:
231 args, kw = fix(args, kw, sig)
--> 232 return caller(func, *(extras + args), **kw)
233 fun.__name__ = func.__name__
234 fun.__doc__ = func.__doc__
~/opt/anaconda3/envs/py310/lib/python3.10/site-packages/IPython/core/magic.py in <lambda>(f, *a, **k)
185 # but it's overkill for just that one bit of state.
186 def magic_deco(arg):
--> 187 call = lambda f, *a, **k: f(*a, **k)
188
189 if callable(arg):
/Applications/Stata/utilities/pystata/ipython/stpymagic.py in stata(self, line, cell, local_ns)
274 _stata.run(cell, quietly=True, inline=_config.stconfig['grshow'])
275 else:
--> 276 _stata.run(cell, quietly=False, inline=_config.stconfig['grshow'])
277
278 if '-gw' in args or '-gh' in args:
/Applications/Stata/utilities/pystata/stata.py in run(cmd, quietly, echo, inline)
323 _stata_wrk2("qui include " + tmpf, None, False, 1)
324 else:
--> 325 _stata_wrk2("include " + tmpf, None, False, 1)
326
327 if inline:
/Applications/Stata/utilities/pystata/stata.py in _stata_wrk2(cmd, real_cmd, colon, mode)
114 err = callback[0]
115 callback.clear()
--> 116 raise SystemError(err)
117 except KeyboardInterrupt:
118 outputter.done()
SystemError: <istmt>: 3200 conformability error
r(3200);
r(3200);
Hi @gcasamat, is this still an issue with the current version of the ado? Do you have a minimum working example of this to see if it is indeed a bug that needs addressing? It seems that the call above is made from Python, so it would be useful to know if this is something to do with Python integration, or sdid natively in Stata.
Hi. Thanks for your reply. I do not have access to the code right now but as far as I remember I got the same error when running it in Stata directly. I will check again with the current version of the ado asap.
Thanks very much! If we know there is an issue here and the characteristics of the data, we will assign it to one of us to aim to resolve. If the error still occurs and you could let us know what version of Stata you are using, that would be v useful!
Hi, I just updated the sdid package and I now get the following error:
synthdid(): 3200 conformability error
<istmt>: - function returned error
Before updating the package, everything was working fine. After checking, it seems to be related to the use of the option vce(jackknife) (it also breaks down with the example code you provide on the sdid webpage) Thanks for your help.