Hello, I've got such an error "data.frame(dict, psspec, weightspec, stringsAsFactors = F) : arguments imply differing number of rows: 5, 4". Can you please explain it? Mayde there're some required characteristics for the covariates, treatment and id variables?
I perform PSM on a dataset of binary and ordinal variables. Maybe this is the problem?
There's also an error "Missing values in dataset" though all the used variables have no such values
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
I think this might have to do with missing values. The program checks
whether there are any missing values in the WHOLE dataset, not just
the variables in the analysis (we will probably change this behavior
in future updates). Try to get rid of those first (e.g., by creating a
subset of variables that are needed for the analysis), and rerun.
Hope that helps,
FT
Hello, I've got such an error "data.frame(dict, psspec, weightspec,
stringsAsFactors = F) : arguments imply differing number of rows: 5, 4". Can
you please explain it? Mayde there're some required characteristics for the
covariates, treatment and id variables?
I perform PSM on a dataset of binary and ordinal variables. Maybe this is
the problem?
There's also an error "Missing values in dataset" though all the used
variables have no such values
Thank you very much! Now everything works properly. The last question: how can export or create a new dataset of matched units?. I've read your manual about ps and psweights automatically created after running PSM. But I don't clearlyy understand how to create matched dataset using these new variables. Can you please help me?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
By default you will get a matched dataset in long format (where each
unit has one row).
If you want to know the matched units go to "Output Datasets" in the
"Plot and Output..." dialogue, and request a wide file (two or more
matched units per row).
Thank you very much! Now everything works properly. The last question: how
can export or create a new dataset of matched units?. I've read your manual
about ps and psweights automatically created after running PSM. But I don't
clearlyy understand how to create matched dataset using these new variables.
Can you please help me?
Hello, Mr Thoemmes! Sorry for asking again. I've recently installed your extension, neccessary R packages on my colleague's PC and used SPSS_EXTENSIONS_PATH to specify the directory for the spe file. . The extension is displayed in the SPSS working panel as a dialog window and R plugin is also working. But when I try to run th PSM there's an error:
PSMATCHING3
/VARS
ID = затраты_техинноваации_совокупный
TREAT = Охватмер
COVS = конкур_рус_совокупный
/MATCHIT
MATCH=NEAREST
EST =LOGIT
DISCARD = NONE
MORDER = LARGEST
RATIO = 1
CALIPER = 0
/PLOT RESOLUTION = 96.
Ошибка: неправильный ввод '.Library = c('C:/Users/kuzyk/Desktop/PSMATCHING3/расширение','C:/PROGRA~2/IBM/SPSS/STATIS~1/20/extensions',.Library)
' в 'utf8towcs'
Ошибка в library(PSMATCHING3, lib.loc = .Library) :
нет пакета под названием 'PSMATCHING3'
Ошибка: не могу найти функцию "Run"
How can I fix it? Thank you in advance
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello Nikolay,
I am not sure what the error messages in cyrillic mean - maybe that it
is not finding a specific path or library. Maybe if you translate, I
might be able to help.
FT
Hello, Mr Thoemmes! Sorry for asking again. I've recently installed your
extension, neccessary R packages on my colleague's PC and used
SPSS_EXTENSIONS_PATH to specify the directory for the spe file. . The
extension is displayed in the SPSS working panel as a dialog window and R
plugin is also working. But when I try to run th PSM there's an error:
PSMATCHING3
/VARS
ID = затраты_техинноваации_совокупный
TREAT = Охватмер
COVS = конкур_рус_совокупный
/MATCHIT
MATCH=NEAREST
EST =LOGIT
DISCARD = NONE
MORDER = LARGEST
RATIO = 1
CALIPER = 0
/PLOT RESOLUTION = 96.
Ошибка: неправильный ввод '.Library =
c('C:/Users/kuzyk/Desktop/PSMATCHING3/расширение','C:/PROGRA~2/IBM/SPSS/STATIS~1/20/extensions',.Library)
' в 'utf8towcs'
Ошибка в library(PSMATCHING3, lib.loc = .Library) :
нет пакета под названием 'PSMATCHING3'
Ошибка: не могу найти функцию "Run"
Hello, I've got such an error "data.frame(dict, psspec, weightspec, stringsAsFactors = F) : arguments imply differing number of rows: 5, 4". Can you please explain it? Mayde there're some required characteristics for the covariates, treatment and id variables?
I perform PSM on a dataset of binary and ordinal variables. Maybe this is the problem?
There's also an error "Missing values in dataset" though all the used variables have no such values
Hi,
I think this might have to do with missing values. The program checks
whether there are any missing values in the WHOLE dataset, not just
the variables in the analysis (we will probably change this behavior
in future updates). Try to get rid of those first (e.g., by creating a
subset of variables that are needed for the analysis), and rerun.
Hope that helps,
FT
On Wed, Apr 8, 2015 at 7:55 AM, Nikolay Zudin nikzud@users.sf.net wrote:
Thank you very much! Now everything works properly. The last question: how can export or create a new dataset of matched units?. I've read your manual about ps and psweights automatically created after running PSM. But I don't clearlyy understand how to create matched dataset using these new variables. Can you please help me?
Hello,
By default you will get a matched dataset in long format (where each
unit has one row).
If you want to know the matched units go to "Output Datasets" in the
"Plot and Output..." dialogue, and request a wide file (two or more
matched units per row).
HTH,
FT
On Mon, Apr 13, 2015 at 7:01 AM, Nikolay Zudin nikzud@users.sf.net wrote:
Perfect! Thank you very much for patience in answering questions and such a useful tool!
Hello, Mr Thoemmes! Sorry for asking again. I've recently installed your extension, neccessary R packages on my colleague's PC and used SPSS_EXTENSIONS_PATH to specify the directory for the spe file. . The extension is displayed in the SPSS working panel as a dialog window and R plugin is also working. But when I try to run th PSM there's an error:
PSMATCHING3
/VARS
ID = затраты_техинноваации_совокупный
TREAT = Охватмер
COVS = конкур_рус_совокупный
/MATCHIT
MATCH=NEAREST
EST =LOGIT
DISCARD = NONE
MORDER = LARGEST
RATIO = 1
CALIPER = 0
/PLOT RESOLUTION = 96.
Ошибка: неправильный ввод '.Library = c('C:/Users/kuzyk/Desktop/PSMATCHING3/расширение','C:/PROGRA~2/IBM/SPSS/STATIS~1/20/extensions',.Library)
' в 'utf8towcs'
Ошибка в library(PSMATCHING3, lib.loc = .Library) :
нет пакета под названием 'PSMATCHING3'
Ошибка: не могу найти функцию "Run"
How can I fix it? Thank you in advance
Hello Nikolay,
I am not sure what the error messages in cyrillic mean - maybe that it
is not finding a specific path or library. Maybe if you translate, I
might be able to help.
FT
On Fri, Apr 17, 2015 at 6:46 AM, Nikolay Zudin nikzud@users.sf.net wrote:
Sure!
PSMATCHING3
/VARS
ID = var 1
TREAT = var 2
COVS = var 3
/MATCHIT
MATCH=NEAREST
EST =LOGIT
DISCARD = NONE
MORDER = LARGEST
RATIO = 1
CALIPER = 0
/PLOT RESOLUTION = 96.
Error: invalid input '.Library = c('C:/Users/kuzyk/Desktop/PSMATCHING3/расширение','C:/PROGRA~2/IBM/SPSS/STATIS~1/20/extensions',.Library)
' in 'utf8towcs'
Error in library(PSMATCHING3, lib.loc = .Library) :
no package named 'PSMATCHING3'
Error: can't find command "Run"
So this means that the installation of the extension bundle did not
succeed. Are you using the latest version 3.04 from sourceforge?
Felix
On Tue, Apr 21, 2015 at 5:21 AM, Nikolay Zudin nikzud@users.sf.net wrote:
3.04? Is that a mistype or is there a unreleased version? 3.03 is the lastest version on Sourceforge.
Any news about your book about propensity score?
Thanks for maintaining this fine piece of software.
Last edit: Lorio 2015-05-21
Sorry, yes 3.04 is not released yet, but soon. Book is in the making,
but will take a bit longer.
On Thu, May 21, 2015 at 3:17 PM, Lorio lorio@users.sf.net wrote: