Impacts code after gstslshet

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Impacts code after gstslshet

R-sig-geo mailing list
Hi,
 
My name is Germán Izón. I am currently working on a project in which I am estimating a SARAR specification using the gstslshet function. After estimating this model I tried estimating the marginal effects using the impacts function but I get the following error: “Error in spdep::intImpacts(rho = rho, beta = beta, P = P, n = n, mu = mu,  : argument "evalues" is missing, with no default”. I get the same message using the columbus data. I do not get this error after estimating a spatial lag model (e.g., lagsarlm). I would truly appreciate any help with this. Thank you!
 
Germán

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Reply | Threaded
Open this post in threaded view
|

Re: Impacts code after gstslshet

Roger Bivand
Administrator
On Fri, 4 Jan 2019, Veron M. Izon via R-sig-Geo wrote:

> Hi,
>
> My name is Germán Izón. I am currently working on a project in which I
> am estimating a SARAR specification using the gstslshet function. After
> estimating this model I tried estimating the marginal effects using the
> impacts function but I get the following error: “Error in
> spdep::intImpacts(rho = rho, beta = beta, P = P, n = n, mu = mu, :
> argument "evalues" is missing, with no default”. I get the same message
> using the columbus data. I do not get this error after estimating a
> spatial lag model (e.g., lagsarlm). I would truly appreciate any help
> with this. Thank you!
Had you provided the output of sessionInfo(), it would be easier to say.
My guess is that you are using sphet 1.7, and spdep 0.8-1. The use of
eigenvalues in impacts measures was presented by Gianfranco Piras at the
Spatial Econometrics meeting in June 2018, and implemented in spdep and
sphet, but sphet has not been released since then. I just bumped the
version number of sphet on R-Forge to 1.7-2, and hope that it will let you
run install.packages("sphet", repos="http://R-Forge.R-project.org"), which
is not possible at the moment (see
https://r-forge.r-project.org/R/?group_id=477). If you can install the
development version of sphet from source, check it out by:

svn checkout svn://svn.r-forge.r-project.org/svnroot/sphet/

and run R CMD build sphet, then R CMD INSTALL that .tar.gz file.

Roger

>
> Germán
>
> _______________________________________________
> R-sig-Geo mailing list
> [hidden email]
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>

--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: [hidden email]
https://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Roger Bivand
Department of Economics
Norwegian School of Economics
Helleveien 30
N-5045 Bergen, Norway
Reply | Threaded
Open this post in threaded view
|

Re: Impacts code after gstslshet

Roger Bivand
Administrator
Run install.packages("sphet", repos="http://R-Forge.R-project.org") for
Linux/OSX source install or Windows binary install of the development
version of sphet, and please report back if this resolves the problem.

Roger

On Sat, 5 Jan 2019, Roger Bivand wrote:

> On Fri, 4 Jan 2019, Veron M. Izon via R-sig-Geo wrote:
>
>>  Hi,
>>
>>  My name is Germán Izón. I am currently working on a project in which I am
>>  estimating a SARAR specification using the gstslshet function. After
>>  estimating this model I tried estimating the marginal effects using the
>>  impacts function but I get the following error: “Error in
>>  spdep::intImpacts(rho = rho, beta = beta, P = P, n = n, mu = mu, :
>>  argument "evalues" is missing, with no default”. I get the same message
>>  using the columbus data. I do not get this error after estimating a
>>  spatial lag model (e.g., lagsarlm). I would truly appreciate any help with
>>  this. Thank you!
>
> Had you provided the output of sessionInfo(), it would be easier to say. My
> guess is that you are using sphet 1.7, and spdep 0.8-1. The use of
> eigenvalues in impacts measures was presented by Gianfranco Piras at the
> Spatial Econometrics meeting in June 2018, and implemented in spdep and
> sphet, but sphet has not been released since then. I just bumped the version
> number of sphet on R-Forge to 1.7-2, and hope that it will let you run
> install.packages("sphet", repos="http://R-Forge.R-project.org"), which is not
> possible at the moment (see https://r-forge.r-project.org/R/?group_id=477).
> If you can install the development version of sphet from source, check it out
> by:
>
> svn checkout svn://svn.r-forge.r-project.org/svnroot/sphet/
>
> and run R CMD build sphet, then R CMD INSTALL that .tar.gz file.
>
> Roger
>
>>
>>  Germán
>>
>>  _______________________________________________
>>  R-sig-Geo mailing list
>>  [hidden email]
>>  https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>>
>
>
--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: [hidden email]
https://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Roger Bivand
Department of Economics
Norwegian School of Economics
Helleveien 30
N-5045 Bergen, Norway
Reply | Threaded
Open this post in threaded view
|

Re: Impacts code after gstslshet

R-sig-geo mailing list
Hi Roger,

Thank you very much for your quick response and help. It did work.

Best,
Germán

On 1/5/19, 7:48 AM, "Roger Bivand" <[hidden email]> wrote:

    Run install.packages("sphet", repos="http://R-Forge.R-project.org") for
    Linux/OSX source install or Windows binary install of the development
    version of sphet, and please report back if this resolves the problem.
   
    Roger
   
    On Sat, 5 Jan 2019, Roger Bivand wrote:
   
    > On Fri, 4 Jan 2019, Veron M. Izon via R-sig-Geo wrote:
    >
    >>  Hi,
    >>
    >>  My name is Germán Izón. I am currently working on a project in which I am
    >>  estimating a SARAR specification using the gstslshet function. After
    >>  estimating this model I tried estimating the marginal effects using the
    >>  impacts function but I get the following error: “Error in
    >>  spdep::intImpacts(rho = rho, beta = beta, P = P, n = n, mu = mu, :
    >>  argument "evalues" is missing, with no default”. I get the same message
    >>  using the columbus data. I do not get this error after estimating a
    >>  spatial lag model (e.g., lagsarlm). I would truly appreciate any help with
    >>  this. Thank you!
    >
    > Had you provided the output of sessionInfo(), it would be easier to say. My
    > guess is that you are using sphet 1.7, and spdep 0.8-1. The use of
    > eigenvalues in impacts measures was presented by Gianfranco Piras at the
    > Spatial Econometrics meeting in June 2018, and implemented in spdep and
    > sphet, but sphet has not been released since then. I just bumped the version
    > number of sphet on R-Forge to 1.7-2, and hope that it will let you run
    > install.packages("sphet", repos="http://R-Forge.R-project.org"), which is not
    > possible at the moment (see https://r-forge.r-project.org/R/?group_id=477).
    > If you can install the development version of sphet from source, check it out
    > by:
    >
    > svn checkout svn://svn.r-forge.r-project.org/svnroot/sphet/
    >
    > and run R CMD build sphet, then R CMD INSTALL that .tar.gz file.
    >
    > Roger
    >
    >>
    >>  Germán
    >>
    >>  _______________________________________________
    >>  R-sig-Geo mailing list
    >>  [hidden email]
    >>  https://stat.ethz.ch/mailman/listinfo/r-sig-geo
    >>
    >
    >
   
    --
    Roger Bivand
    Department of Economics, Norwegian School of Economics,
    Helleveien 30, N-5045 Bergen, Norway.
    voice: +47 55 95 93 55; e-mail: [hidden email]
    https://orcid.org/0000-0003-2392-6140
    https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo