what is the difference between using function predict() and coef() in prediction

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

what is the difference between using function predict() and coef() in prediction

tony333
i do not know what is the difference between predict() and coef()
i use the two function give me different result
zz = predict(xy.lm,list(T8=T8))
ss = coef(xy.lm)[1]+(coef(xy.lm)[2])*T8
where is t8 is the data used in prediction did not use in the fitting in training sample where every sample is 15 value
Reply | Threaded
Open this post in threaded view
|

Re: what is the difference between using function predict() and coef() in prediction

PIKAL Petr
Hi

Did you find any difference? The results shall be same (with only rounding
error). Can you show us some example where you get substantial difference?

Regards
Petr

>
> i do not know what is the difference between predict() and coef()
> i use the two function give me different result
> zz = predict(xy.lm,list(T8=T8))
> ss = coef(xy.lm)[1]+(coef(xy.lm)[2])*T8
> where is t8 is the data used in prediction did not use in the fitting in
> training sample where every sample is 15 value
>
>
> --
> View this message in context: http://r.789695.n4.nabble.com/what-is-the-
> difference-between-using-function-predict-and-coef-in-prediction-
> tp4173961p4173961.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> [hidden email] mailing list
> 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.

______________________________________________
[hidden email] mailing list
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.
Reply | Threaded
Open this post in threaded view
|

Re: what is the difference between using function predict() and coef() in prediction

tony333

X8 = c(0.42808332,  0.14058333,  0.30558333,  0.09558333,  0.01808333,
-0.09191666, -0.11441666, -0.12941666,  0.13808333, -0.31691666,  0.25308333
,-0.20941666 ,0.02808333, -0.04441667, -0.43691666)  

xy.lm =  lm(Y~X8)
z = predict(xy.lm,list(X8=X8))
sz = coef(xy.lm)[1]+(coef(xy.lm)[2])*X8
is the same result in this step
but when use the prediction as
T8=c(-0.05691667, -0.09441666 , 0.31308333 ,-0.59941665 ,-0.33941666
,-0.64441665,-0.55691665 ,-0.01191667  ,0.22558333, -0.30691666  ,0.97808331
, 0.49808332,-0.30941666  ,0.41808332 , 0.42308332)

zz = predict(xy.lm,list(T8=T8))
ss = coef(xy.lm)[1]+(coef(xy.lm)[2])*T8
here gives me different  value and  zz = z = sz i do not know the reason for
that
what i understand about predict() act exactly like using coef() but now do
not know  the difference
thank you for respond  

--
View this message in context: http://r.789695.n4.nabble.com/what-is-the-difference-between-using-function-predict-and-coef-in-prediction-tp4173961p4180577.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
[hidden email] mailing list
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.
Reply | Threaded
Open this post in threaded view
|

Re: what is the difference between using function predict() and coef() in prediction

tony333
In reply to this post by PIKAL Petr
X8 = c(0.42808332,  0.14058333,  0.30558333,  0.09558333,  0.01808333, -0.09191666, -0.11441666, -0.12941666,  0.13808333, -0.31691666,  0.25308333 ,-0.20941666 ,0.02808333, -0.04441667, -0.43691666)  

Y =c(370.63333  , 887.63333  ,3610.88333 ,  435.13333 , 1261.38333 , -741.11667,-3231.36667  ,-708.36667  ,-646.36667, -3652.86667 ,-2196.11667 ,-5552.61667,-92.11667, -1635.86667 ,-3264.61667)

xy.lm =  lm(Y~X8)
z = predict(xy.lm,list(X8=X8))
sz = coef(xy.lm)[1]+(coef(xy.lm)[2])*X8
is the same result in this step
but when use the prediction as
T8=c(-0.05691667, -0.09441666 , 0.31308333 ,-0.59941665 ,-0.33941666 ,-0.64441665,-0.55691665 ,-0.01191667  ,0.22558333, -0.30691666  ,0.97808331 , 0.49808332,-0.30941666  ,0.41808332 , 0.42308332)

zz = predict(xy.lm,list(T8=T8))
ss = coef(xy.lm)[1]+(coef(xy.lm)[2])*T8
here gives me different  value and  zz = z = sz i do not know the reason for that
what i understand about predict() act exactly like using coef() but now do not know  the difference
thank you for respond  
Reply | Threaded
Open this post in threaded view
|

Re: what is the difference between using function predict() and coef() in prediction

David Winsemius
In reply to this post by tony333

On Dec 10, 2011, at 11:48 AM, tony333 wrote:

>
> X8 = c(0.42808332,  0.14058333,  0.30558333,  0.09558333,  0.01808333,
> -0.09191666, -0.11441666, -0.12941666,  0.13808333, -0.31691666,  
> 0.25308333
> ,-0.20941666 ,0.02808333, -0.04441667, -0.43691666)
>
> xy.lm =  lm(Y~X8)
> z = predict(xy.lm,list(X8=X8))
> sz = coef(xy.lm)[1]+(coef(xy.lm)[2])*X8
> is the same result in this step
> but when use the prediction as
> T8=c(-0.05691667, -0.09441666 , 0.31308333 ,-0.59941665 ,-0.33941666
> ,-0.64441665,-0.55691665 ,-0.01191667  ,0.22558333, -0.30691666  ,
> 0.97808331
> , 0.49808332,-0.30941666  ,0.41808332 , 0.42308332)
>
> zz = predict(xy.lm,list(T8=T8))

The first "T8" should be..^^ .... "X8"

 > zz = predict(xy.lm,data.frame(X8=T8))
 > ss = coef(xy.lm)[1]+(coef(xy.lm)[2])*T8

 > identical(unname(zz), ss)
[1] TRUE

> ss = coef(xy.lm)[1]+(coef(xy.lm)[2])*T8
> here gives me different  value and  zz = z = sz i do not know the  
> reason for
> that
> what i understand about predict() act exactly like using coef() but  
> now do
> not know  the difference

If you give predict alist or data.fram whose name is different than  
that of the origianl variable ist will be ignored.

> thank you for respond
>
> --
> View this message in context: http://r.789695.n4.nabble.com/what-is-the-difference-between-using-function-predict-and-coef-in-prediction-tp4173961p4180577.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> [hidden email] mailing list
> 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.

David Winsemius, MD
West Hartford, CT

______________________________________________
[hidden email] mailing list
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.