Hitting the upper bound?

[asheshrambachan]

Hitting the upper bound?

[jonathandroth]

Have you tried e.g.: honest_did(es = es2, e = 7, type="relative_magnitude", grid.ub = 3) if you want to make the grid ub 3

…

On Thu, Jan 25, 2024 at 1:35 PM Elsa Voytas ***@***.***> wrote: Hello! Thank you for the package and for your help! I'm using the did package and conducting sensitivity analysis with the honest_did auxiliary function <https://github.com/pedrohcgs/CS_RR>. At some level of $\bar{M}$, the confidence intervals stop increasing, perhaps because they've reached the plot's upper/lower bound. Here is the code used to produce the result, with period 7 as an example: att2 <- att_gt(yname = "vote", tname = "time.id", idname = "ID", gname = "firstrep", xformla = ~age + female, data = panel.df, base_period = "universal", control_group = "notyettreated") es2 <- aggte(att2, type = "dynamic", na.rm = T, min_e = -12, max_e = 12) sensitivityresults7 <- honest_did(es = es2, e = 7, type="relative_magnitude") sensitivityresults7$robust_ci <- sensitivityresults7$robust_ci[-1,] createSensitivityPlot_relativeMagnitudes(sensitivityresults7$robust_ci, sensitivityresults7$orig_ci) The output is: image.png (view on web) <https://github.com/asheshrambachan/HonestDiD/assets/32146369/bbb83970-b562-4b49-9106-0adae7a583c6> The data can be found at: https://drive.google.com/drive/folders/1N5HEWIIZSX_EgKzrmX9R9hzcn5fE5zyq?usp=sharing Do you have a suggestion to expand the bounds using this approach? Thank you!! — Reply to this email directly, view it on GitHub <#52>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AE6EXFDRUFG243MOVJGE32DYQKQWVAVCNFSM6AAAAABCK7TYBGVHI2DSMVQWIX3LMV43ASLTON2WKOZSGEYDAOJVGU2TGNI> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[elsavoytas]

Awesome! Thank you! That helped!

sensitivityresults7 <- honest_did(es = es2, 
                   e = 7,
                   type="relative_magnitude", grid.ub = 3, grid.lb = -1.5)

sensitivityresults7$robust_ci <- sensitivityresults7$robust_ci[-1,]

createSensitivityPlot_relativeMagnitudes(sensitivityresults7$robust_ci,
                                                    sensitivityresults7$orig_ci)

yielded the plot below.

I'm not sure why the two final CIs have the same upper bound and the 4th and 5th from the right have the same lower bound? Experimenting with slightly different ub and lb generates the same behavior but for different CIs.

[jonathandroth]

Great. Likely because honestdid uses a discrete grid so the upper bounds can be the same if the upper grid points are the same. You can also pass num.gridpoints to change grid resolution

…

On Thu, Jan 25, 2024, 5:22 PM Elsa Voytas ***@***.***> wrote: Awesome! Thank you! That helped! sensitivityresults7 <- honest_did(es = es2, e = 7, type="relative_magnitude", grid.ub = 3, grid.lb = -1.5) sensitivityresults7$robust_ci <- sensitivityresults7$robust_ci[-1,] createSensitivityPlot_relativeMagnitudes(sensitivityresults7$robust_ci, sensitivityresults7$orig_ci) yielded the plot below. image.png (view on web) <https://github.com/asheshrambachan/HonestDiD/assets/32146369/93284574-3bd0-43d6-8fed-9e015cbc5f13> I'm not sure why the two final CIs have the same upper bound and the 4th and 5th from the right have the same lower bound? Experimenting with slightly different ub and lb generates the same behavior but for different CIs. — Reply to this email directly, view it on GitHub <#52 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AE6EXFB4JQVUGK6SO2UIWF3YQLLKBAVCNFSM6AAAAABCK7TYBGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMJRGA4TIMRQGE> . You are receiving this because you commented.Message ID: ***@***.***>

[jonathandroth]

If you are happy with the current grid resolution I don't think it's a problem per we if two of the bounds match

…

On Thu, Jan 25, 2024, 5:35 PM Jonathan Roth ***@***.***> wrote: Great. Likely because honestdid uses a discrete grid so the upper bounds can be the same if the upper grid points are the same. You can also pass num.gridpoints to change grid resolution On Thu, Jan 25, 2024, 5:22 PM Elsa Voytas ***@***.***> wrote: > Awesome! Thank you! That helped! > > sensitivityresults7 <- honest_did(es = es2, > e = 7, > type="relative_magnitude", grid.ub = 3, grid.lb = -1.5) > > sensitivityresults7$robust_ci <- sensitivityresults7$robust_ci[-1,] > > createSensitivityPlot_relativeMagnitudes(sensitivityresults7$robust_ci, > sensitivityresults7$orig_ci) > > yielded the plot below. > image.png (view on web) > <https://github.com/asheshrambachan/HonestDiD/assets/32146369/93284574-3bd0-43d6-8fed-9e015cbc5f13> > > I'm not sure why the two final CIs have the same upper bound and the 4th > and 5th from the right have the same lower bound? Experimenting with > slightly different ub and lb generates the same behavior but for different > CIs. > > — > Reply to this email directly, view it on GitHub > <#52 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AE6EXFB4JQVUGK6SO2UIWF3YQLLKBAVCNFSM6AAAAABCK7TYBGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMJRGA4TIMRQGE> > . > You are receiving this because you commented.Message ID: > ***@***.***> >

[elsavoytas]

Agreed! Thanks for your help. I'm all set.