Issue bootstrapping when L1 Ratio % = 0

[lmgibson]

No issue when running bootstrap = FALSE.

Input:

tjbal(Y ~ D, data = as.data.frame(d),
index = c("id","Time"), kernel = FALSE, demean = FALSE, bootstrap = TRUE, nboots = 100)

Output:

Seek balance on: MedCost1, MedCost2, MedCost3, MedCost4, MedCost5, MedCost6, MedCost7

Optimization: [1] "Trying 1 dims of K; distance at 0.12504" [1] "Trying 2 dims of K; distance at 0.10882" [1] "Trying 3 dims of K; distance at 0.09303" [1] "Trying 4 dims of K; distance at 0.0936" [1] "Trying 5 dims of K; distance at 0.09304" [1] "Trying 6 dims of K; distance at 0.10682" [1] "Trying 7 dims of K; distance at 0" sigma* = 1; L1.ratio% = 0.000; num.dims = 7

Bootstrapping...Parallel computing...Error in { : task 1 failed - "system is computationally singular: reciprocal condition number = 2.12835e-18"

[xuyiqing]

Dear Landon,

Would it be possible if you can send me a sample dataset such that I can replicate the error?

Thanks, Yiqing

On Tue, Oct 23, 2018 at 9:43 AM Landon Gibson notifications@github.com wrote:

No issue when running bootstrap = FALSE.

Input:

tjbal(Y ~ D, data = as.data.frame(d), index = c("id","Time"), kernel = FALSE, demean = FALSE, bootstrap = TRUE, nboots = 100)

Output:

Seek balance on: MedCost1, MedCost2, MedCost3, MedCost4, MedCost5, MedCost6, MedCost7

Optimization: [1] "Trying 1 dims of K; distance at 0.12504" [1] "Trying 2 dims of K; distance at 0.10882" [1] "Trying 3 dims of K; distance at 0.09303" [1] "Trying 4 dims of K; distance at 0.0936" [1] "Trying 5 dims of K; distance at 0.09304" [1] "Trying 6 dims of K; distance at 0.10682" [1] "Trying 7 dims of K; distance at 0" sigma* = 1; L1.ratio% = 0.000; num.dims = 7

Bootstrapping...Parallel computing...Error in { : task 1 failed - "system is computationally singular: reciprocal condition number = 2.12835e-18"

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/xuyiqing/tjbal/issues/2, or mute the thread https://github.com/notifications/unsubscribe-auth/AHT1GDvXzn-j3v2gIAdrAXw84CBzR2Hqks5un0dMgaJpZM4X2Ame .

-- Yiqing Xu

Assistant Professor Department of Political Science University of California, San Diego http://yiqingxu.org/

[lmgibson]

Hi Dr. Xu,

Thanks for the quick response. Attached is some code and a sample dataset saved out as an RDS file.

Also, as I was setting up the replication code and making sure all my packages were up to date I noticed the following error when trying to install the tjbal package:

devtools::install_github('xuyiqing/tjbal’)

"Error in FUN(X[[i]], ...) : Invalid comparison operator in dependency: >= “

Let me know if you need anything else.

• Landon On Oct 23, 2018, 12:34 PM -0700, Yiqing Xu notifications@github.com, wrote:

  Dear Landon,

  Would it be possible if you can send me a sample dataset such that I can replicate the error?

  Thanks, Yiqing

  On Tue, Oct 23, 2018 at 9:43 AM Landon Gibson notifications@github.com wrote:

  No issue when running bootstrap = FALSE.

  Input:

  tjbal(Y ~ D, data = as.data.frame(d), index = c("id","Time"), kernel = FALSE, demean = FALSE, bootstrap = TRUE, nboots = 100)

  Output:

  Seek balance on: MedCost1, MedCost2, MedCost3, MedCost4, MedCost5, MedCost6, MedCost7

  Optimization: [1] "Trying 1 dims of K; distance at 0.12504" [1] "Trying 2 dims of K; distance at 0.10882" [1] "Trying 3 dims of K; distance at 0.09303" [1] "Trying 4 dims of K; distance at 0.0936" [1] "Trying 5 dims of K; distance at 0.09304" [1] "Trying 6 dims of K; distance at 0.10682" [1] "Trying 7 dims of K; distance at 0" sigma* = 1; L1.ratio% = 0.000; num.dims = 7

  Bootstrapping...Parallel computing...Error in { : task 1 failed - "system is computationally singular: reciprocal condition number = 2.12835e-18"

  — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/xuyiqing/tjbal/issues/2, or mute the thread https://github.com/notifications/unsubscribe-auth/AHT1GDvXzn-j3v2gIAdrAXw84CBzR2Hqks5un0dMgaJpZM4X2Ame .

  -- Yiqing Xu

  Assistant Professor Department of Political Science University of California, San Diego http://yiqingxu.org/ — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

[xuyiqing]

I noticed that. It is a bug of the devtools packages. I'll look into your problem.

On Tue, Oct 23, 2018 at 1:12 PM Landon Gibson notifications@github.com wrote:

Hi Dr. Xu,

Thanks for the quick response. Attached is some code and a sample dataset saved out as an RDS file.

Also, as I was setting up the replication code and making sure all my packages were up to date I noticed the following error when trying to install the tjbal package:

devtools::install_github('xuyiqing/tjbal’)

"Error in FUN(X[[i]], ...) : Invalid comparison operator in dependency: >= “

Let me know if you need anything else.

• Landon On Oct 23, 2018, 12:34 PM -0700, Yiqing Xu notifications@github.com, wrote:

  Dear Landon,

  Would it be possible if you can send me a sample dataset such that I can replicate the error?

  Thanks, Yiqing

  On Tue, Oct 23, 2018 at 9:43 AM Landon Gibson notifications@github.com wrote:

  No issue when running bootstrap = FALSE.

  Input:

  tjbal(Y ~ D, data = as.data.frame(d), index = c("id","Time"), kernel = FALSE, demean = FALSE, bootstrap = TRUE, nboots = 100)

  Output:

  Seek balance on: MedCost1, MedCost2, MedCost3, MedCost4, MedCost5, MedCost6, MedCost7

  Optimization: [1] "Trying 1 dims of K; distance at 0.12504" [1] "Trying 2 dims of K; distance at 0.10882" [1] "Trying 3 dims of K; distance at 0.09303" [1] "Trying 4 dims of K; distance at 0.0936" [1] "Trying 5 dims of K; distance at 0.09304" [1] "Trying 6 dims of K; distance at 0.10682" [1] "Trying 7 dims of K; distance at 0" sigma* = 1; L1.ratio% = 0.000; num.dims = 7

  Bootstrapping...Parallel computing...Error in { : task 1 failed - "system is computationally singular: reciprocal condition number = 2.12835e-18"

  — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/xuyiqing/tjbal/issues/2, or mute the thread < https://github.com/notifications/unsubscribe-auth/AHT1GDvXzn-j3v2gIAdrAXw84CBzR2Hqks5un0dMgaJpZM4X2Ame

  .

  -- Yiqing Xu

  Assistant Professor Department of Political Science University of California, San Diego http://yiqingxu.org/ — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/xuyiqing/tjbal/issues/2#issuecomment-432401127, or mute the thread https://github.com/notifications/unsubscribe-auth/AHT1GMA1cH6h4NVbQkJYtsGkM8ef8qY-ks5un3gqgaJpZM4X2Ame .

-- Yiqing Xu

Assistant Professor Department of Political Science University of California, San Diego http://yiqingxu.org/

[xuyiqing]

Hi Landon,

Could you send an email (with attachment) directly to my email account? yiqingxu@ucsd.edu

Thanks!

Best, Yiqing

On Tue, Oct 23, 2018 at 1:12 PM Landon Gibson notifications@github.com wrote:

Hi Dr. Xu,

Thanks for the quick response. Attached is some code and a sample dataset saved out as an RDS file.

Also, as I was setting up the replication code and making sure all my packages were up to date I noticed the following error when trying to install the tjbal package:

devtools::install_github('xuyiqing/tjbal’)

"Error in FUN(X[[i]], ...) : Invalid comparison operator in dependency: >= “

Let me know if you need anything else.

• Landon On Oct 23, 2018, 12:34 PM -0700, Yiqing Xu notifications@github.com, wrote:

  Dear Landon,

  Would it be possible if you can send me a sample dataset such that I can replicate the error?

  Thanks, Yiqing

  On Tue, Oct 23, 2018 at 9:43 AM Landon Gibson notifications@github.com wrote:

  No issue when running bootstrap = FALSE.

  Input:

  tjbal(Y ~ D, data = as.data.frame(d), index = c("id","Time"), kernel = FALSE, demean = FALSE, bootstrap = TRUE, nboots = 100)

  Output:

  Seek balance on: MedCost1, MedCost2, MedCost3, MedCost4, MedCost5, MedCost6, MedCost7

  Optimization: [1] "Trying 1 dims of K; distance at 0.12504" [1] "Trying 2 dims of K; distance at 0.10882" [1] "Trying 3 dims of K; distance at 0.09303" [1] "Trying 4 dims of K; distance at 0.0936" [1] "Trying 5 dims of K; distance at 0.09304" [1] "Trying 6 dims of K; distance at 0.10682" [1] "Trying 7 dims of K; distance at 0" sigma* = 1; L1.ratio% = 0.000; num.dims = 7

  Bootstrapping...Parallel computing...Error in { : task 1 failed - "system is computationally singular: reciprocal condition number = 2.12835e-18"

  — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/xuyiqing/tjbal/issues/2, or mute the thread < https://github.com/notifications/unsubscribe-auth/AHT1GDvXzn-j3v2gIAdrAXw84CBzR2Hqks5un0dMgaJpZM4X2Ame

  .

  -- Yiqing Xu

  Assistant Professor Department of Political Science University of California, San Diego http://yiqingxu.org/ — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/xuyiqing/tjbal/issues/2#issuecomment-432401127, or mute the thread https://github.com/notifications/unsubscribe-auth/AHT1GMA1cH6h4NVbQkJYtsGkM8ef8qY-ks5un3gqgaJpZM4X2Ame .

-- Yiqing Xu

Assistant Professor Department of Political Science University of California, San Diego http://yiqingxu.org/

[xuyiqing]

Bug is fixed.