R help - Apr 2024 - Synthetic Control Method

Good Morning

 

I want to perform a synthetic control method with R. For this purpose, I
created the following code:

 

# Re-load packages

library(Synth)

library(readxl)

 

# Pfadeinstellung Excel-Blatt

excel_file_path <-
("C:\\Users\\xxxxx\\Desktop\\DATA_INVESTMENTVOLUMEN_FOR_R_WITHOUT_NA.xlsx")

 

# Load the Excel file

INVESTMENTVOLUME <- read_excel(excel_file_path)

 

# Anzeigen des gesamten Dataframes

print(INVESTMENTVOLUME)

 

# Make sure BFS is numeric right before dataprep

INVESTMENTVOLUME$BFS <- as.numeric(INVESTMENTVOLUME$BFS)

 

# running dataprep 

dataprep_out <- dataprep(

  foo = INVESTMENTVOLUME,

  predictors = c("Predictor 1", " Predictor 2", "
Predictor 3" Predictor 4",
" Predictor 5", " Predictor 6"),

  special.predictors = list(list("Special Predictor 1", seq(1, 12, by
1))),

  dependent = "INVESTMENTVOLUME_12_MONTH_AVERAGE",

  unit.variable = "BFS",

  time.variable = "DATE",

  treatment.identifier = XXXX,

  controls.identifier unique(INVESTMENTVOLUME$BFS[-which(INVESTMENTVOLUME$BFS ==
XXXX)]),

  time.predictors.prior =
as.Date("2010-01-01"):as.Date("2017-10-01"),

  time.optimize.ssr =
as.Date("2010-01-01"):as.Date("2017-10-01"),

  time.plot = as.Date("2010-01-01"):as.Date("2024-03-01"),

  unit.names.variable = "BFS"

)

 

synth_out <- synth(

  data.prep.obj = dataprep_out

)

 

I keep getting the same error message. Unfortunately, ChatGPT and solutions
from various forums do not help. My unit variables are all numeric except
one, which is a date and has POSIXct as type.
 
Fehler in dataprep(foo = INVESTMENTVOLUME, predictors = c("Predictor
1",  :

  

 unit.variable not found as numeric variable in foo.

 

I would be very grateful if you could help me with my problem.

 

Thank you in advance for your efforts.

 

Kind regards

Nadja Delliehausen


	[[alternative HTML version deleted]]

Note that your unit.variable and unit.names.variable are indentical. Is
this what you intended?

(I have no idea how the Synth package works).

Bert

On Tue, Apr 16, 2024 at 12:58?AM <nadja.delliehausen at bluewin.ch> wrote:
> Good Morning
>
>
>
> I want to perform a synthetic control method with R. For this purpose, I
> created the following code:
>
>
>
> # Re-load packages
>
> library(Synth)
>
> library(readxl)
>
>
>
> # Pfadeinstellung Excel-Blatt
>
> excel_file_path <-
>
("C:\\Users\\xxxxx\\Desktop\\DATA_INVESTMENTVOLUMEN_FOR_R_WITHOUT_NA.xlsx")
>
>
>
> # Load the Excel file
>
> INVESTMENTVOLUME <- read_excel(excel_file_path)
>
>
>
> # Anzeigen des gesamten Dataframes
>
> print(INVESTMENTVOLUME)
>
>
>
> # Make sure BFS is numeric right before dataprep
>
> INVESTMENTVOLUME$BFS <- as.numeric(INVESTMENTVOLUME$BFS)
>
>
>
> # running dataprep
>
> dataprep_out <- dataprep(
>
>   foo = INVESTMENTVOLUME,
>
>   predictors = c("Predictor 1", " Predictor 2", "
Predictor 3" Predictor
> 4",
> " Predictor 5", " Predictor 6"),
>
>   special.predictors = list(list("Special Predictor 1", seq(1,
12, by > 1))),
>
>   dependent = "INVESTMENTVOLUME_12_MONTH_AVERAGE",
>
>   unit.variable = "BFS",
>
>   time.variable = "DATE",
>
>   treatment.identifier = XXXX,
>
>   controls.identifier >
unique(INVESTMENTVOLUME$BFS[-which(INVESTMENTVOLUME$BFS == XXXX)]),
>
>   time.predictors.prior =
as.Date("2010-01-01"):as.Date("2017-10-01"),
>
>   time.optimize.ssr =
as.Date("2010-01-01"):as.Date("2017-10-01"),
>
>   time.plot =
as.Date("2010-01-01"):as.Date("2024-03-01"),
>
>   unit.names.variable = "BFS"
>
> )
>
>
>
> synth_out <- synth(
>
>   data.prep.obj = dataprep_out
>
> )
>
>
>
> I keep getting the same error message. Unfortunately, ChatGPT and solutions
> from various forums do not help. My unit variables are all numeric except
> one, which is a date and has POSIXct as type.
>
> Fehler in dataprep(foo = INVESTMENTVOLUME, predictors = c("Predictor
1",
> :
>
>
>
>  unit.variable not found as numeric variable in foo.
>
>
>
> I would be very grateful if you could help me with my problem.
>
>
>
> Thank you in advance for your efforts.
>
>
>
> Kind regards
>
> Nadja Delliehausen
>
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
	[[alternative HTML version deleted]]

Hallo Nadja

Similar as Bert I do not know how the function works. From the help page -
synth.data - is used in example. Check if your data structure is consistent
with synth.data by comparing str(synth.data) and str( INVESTMENTVOLUME).
Names of columns should be in both cases match the names in dataprep call
and type of columns should be also same.

Cheers.
Petr



<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Neobsahuje
??dn? viry.www.avast.com
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
<#m_7252664788356473291_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

?t 16. 4. 2024 v 9:58 odes?latel <nadja.delliehausen at bluewin.ch>
napsal:
> Good Morning
>
>
>
> I want to perform a synthetic control method with R. For this purpose, I
> created the following code:
>
>
>
> # Re-load packages
>
> library(Synth)
>
> library(readxl)
>
>
>
> # Pfadeinstellung Excel-Blatt
>
> excel_file_path <-
>
("C:\\Users\\xxxxx\\Desktop\\DATA_INVESTMENTVOLUMEN_FOR_R_WITHOUT_NA.xlsx")
>
>
>
> # Load the Excel file
>
> INVESTMENTVOLUME <- read_excel(excel_file_path)
>
>
>
> # Anzeigen des gesamten Dataframes
>
> print(INVESTMENTVOLUME)
>
>
>
> # Make sure BFS is numeric right before dataprep
>
> INVESTMENTVOLUME$BFS <- as.numeric(INVESTMENTVOLUME$BFS)
>
>
>
> # running dataprep
>
> dataprep_out <- dataprep(
>
>   foo = INVESTMENTVOLUME,
>
>   predictors = c("Predictor 1", " Predictor 2", "
Predictor 3" Predictor
> 4",
> " Predictor 5", " Predictor 6"),
>
>   special.predictors = list(list("Special Predictor 1", seq(1,
12, by > 1))),
>
>   dependent = "INVESTMENTVOLUME_12_MONTH_AVERAGE",
>
>   unit.variable = "BFS",
>
>   time.variable = "DATE",
>
>   treatment.identifier = XXXX,
>
>   controls.identifier >
unique(INVESTMENTVOLUME$BFS[-which(INVESTMENTVOLUME$BFS == XXXX)]),
>
>   time.predictors.prior =
as.Date("2010-01-01"):as.Date("2017-10-01"),
>
>   time.optimize.ssr =
as.Date("2010-01-01"):as.Date("2017-10-01"),
>
>   time.plot =
as.Date("2010-01-01"):as.Date("2024-03-01"),
>
>   unit.names.variable = "BFS"
>
> )
>
>
>
> synth_out <- synth(
>
>   data.prep.obj = dataprep_out
>
> )
>
>
>
> I keep getting the same error message. Unfortunately, ChatGPT and solutions
> from various forums do not help. My unit variables are all numeric except
> one, which is a date and has POSIXct as type.
>
> Fehler in dataprep(foo = INVESTMENTVOLUME, predictors = c("Predictor
1",
> :
>
>
>
>  unit.variable not found as numeric variable in foo.
>
>
>
> I would be very grateful if you could help me with my problem.
>
>
>
> Thank you in advance for your efforts.
>
>
>
> Kind regards
>
> Nadja Delliehausen
>
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
	[[alternative HTML version deleted]]