Re: How to get each bandwidth values on observation points using adaptive kernel in spgwr ?

[Husna]

when i run

res <- gw.adapt(cbind(columbus$x, columbus$y), cbind(columbus$x,
columbus$y), quant=0.5)


there is error message

Error: could not find function "gw.adapt"


why did it happen? 

 
Best Regards,
Lilis Nurul Husna
Institute of Statistics 

Jakarta

        [[alternative HTML version deleted]]


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

[Husna]

when i run


res <- gw.adapt(cbind(columbus$x, columbus$y), cbind(columbus$x,
columbus$y), quant=0.5)


there is error message

Error: could not find function "gw.adapt"


why did it happen? 
 
Best Regards,
Lilis Nurul Husna
Institute of Statistics 

Jakarta
        [[alternative HTML version deleted]]


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

[Barry Rowlingson]

On Mon, Jul 16, 2012 at 10:36 AM, Lilis Husna <[hidden email]> wrote:

> when i run
>
>
> res <- gw.adapt(cbind(columbus$x, columbus$y), cbind(columbus$x,
> columbus$y), quant=0.5)
>
>
> there is error message
>
> Error: could not find function "gw.adapt"
>
>
> why did it happen?

 Who told you about gw.adapt?

 It seems to now be not exported from the spgwr package, but the help
page is still there (Roger: is that a bug? Either you've missed it out
of the exports or you've left in the .Rd file!).

 If you want to use it you can use the triple-colon notation to get an
unexported function from a package:

res = spgwr:::gw.adapt(cbind(columbus$x,columbus$y),cbind(columbus$x,columbus$y),quant=0.5)

 Roger may have removed it for a good reason, and possibly put the
functionality into another function. But the quick fix would be the
triple-colon trick.

 It's also useful if you give package versions with your questions -
here's mine:

> packageDescription("spgwr")$Version
[1] "0.6-14"

Barry

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

[Roger Bivand]

On Mon, 16 Jul 2012, Barry Rowlingson wrote:

> On Mon, Jul 16, 2012 at 10:36 AM, Lilis Husna <[hidden email]> wrote:
>> when i run
>>
>>
>> res <- gw.adapt(cbind(columbus$x, columbus$y), cbind(columbus$x,
>> columbus$y), quant=0.5)
>>
>>
>> there is error message
>>
>> Error: could not find function "gw.adapt"
>>
>>
>> why did it happen?
>
> Who told you about gw.adapt?
>
> It seems to now be not exported from the spgwr package, but the help
> page is still there (Roger: is that a bug? Either you've missed it out
> of the exports or you've left in the .Rd file!).

I hadn't intended to export it since I cannot see that users need to use
it directly, but forgot to remove the help file. I've committed gw.adapt
in the NAMESPACE to R-Forge, and it will be exposed (since a user wants
it) in the next release, or installing from R-Forge directly. For the
currently released package, follow Barry's advice and use the spgwr:::
prefix.

Roger

>
> If you want to use it you can use the triple-colon notation to get an
> unexported function from a package:
>
> res = spgwr:::gw.adapt(cbind(columbus$x,columbus$y),cbind(columbus$x,columbus$y),quant=0.5)
>
> Roger may have removed it for a good reason, and possibly put the
> functionality into another function. But the quick fix would be the
> triple-colon trick.
>
> It's also useful if you give package versions with your questions -
> here's mine:
>
>> packageDescription("spgwr")$Version
> [1] "0.6-14"
>
> Barry
>

--
Roger Bivand
Department of Economics, NHH Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: [phone]; fax [phone]
e-mail: [hidden email]

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

[Husna]

Thank you for all your information.
I got it.

How about estimate each parameters of gwr with R? including, estimate in non sample area?

 
Best Regards,
Lilis Nurul Husna
Institute of Statistics 

Jakarta



________________________________
 Dari: Roger Bivand <[hidden email]>
Kepada: Barry Rowlingson <[hidden email]>

[hidden email]>
Dikirim: Senin, 16 Juli 2012 20:29
Judul: Re: [R-sig-Geo] How to get each bandwidth values on observation points using adaptive kernel in spgwr ?

On Mon, 16 Jul 2012, Barry Rowlingson wrote:


wrote:

>> when i run
>>
>>
>> res <- gw.adapt(cbind(columbus$x, columbus$y), cbind(columbus$x,
>> columbus$y), quant=0.5)
>>
>>
>> there is error message
>>
>> Error: could not find function "gw.adapt"
>>
>>
>> why did it happen?
>
> Who told you about gw.adapt?
>
> It seems to now be not exported from the spgwr package, but the help
> page is still there (Roger: is that a bug? Either you've missed it out
> of the exports or you've left in the .Rd file!).

I hadn't intended to export it since I cannot see that users need to use
it directly, but forgot to remove the help file. I've committed gw.adapt
in the NAMESPACE to R-Forge, and it will be exposed (since a user wants
it) in the next release, or installing from R-Forge directly. For the
currently released package, follow Barry's advice and use the spgwr:::
prefix.

Roger

>
> If you want to use it you can use the triple-colon notation to get an
> unexported function from a package:
>
> res = spgwr:::gw.adapt(cbind(columbus$x,columbus$y),cbind(columbus$x,columbus$y),quant=0.5)
>
> Roger may have removed it for a good reason, and possibly put the
> functionality into another function. But the quick fix would be the
> triple-colon trick.
>
> It's also useful if you give package versions with your questions -
> here's mine:
>
>> packageDescription("spgwr")$Version
> [1] "0.6-14"
>
> Barry
>

--
Roger Bivand
Department of Economics, NHH Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: [phone]; fax [phone]
e-mail: [hidden email]
        [[alternative HTML version deleted]]


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

[Husna]

I want to ask about gwr.morantest in spgwr package.

What is the purpose of doing this test?

Which variabel is tested? residual from global model (OLS), residual from local model (GWR), or the dependent variabel (same as testing spatial autocorrelation)?

Best Regards,
Lilis Nurul Husna
Institute of Statistics r

Jakarta
        [[alternative HTML version deleted]]


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

[Roger Bivand]

On Wed, 25 Jul 2012, Lilis Husna wrote:

>
>
> I want to ask about gwr.morantest in spgwr package.
>
> What is the purpose of doing this test?
>
> Which variabel is tested? residual from global model (OLS), residual
> from local model (GWR), or the dependent variabel (same as testing
> spatial autocorrelation)?

The residual from the local model (GWR).

Roger

>
> Best Regards,
> Lilis Nurul Husna
> Institute of Statistics?r
>
> Jakarta
> [[alternative HTML version deleted]]
>
>

--
Roger Bivand
Department of Economics, NHH Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: [phone]; fax [phone]
e-mail: [hidden email]

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