R help - Dec 2011 - what is the difference between using function predict() and coef() in prediction

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.

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.
> 
> ______________________________________________
> R-help at r-project.org 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.

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.

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  

--
View this message in context:
http://r.789695.n4.nabble.com/what-is-the-difference-between-using-function-predict-and-coef-in-prediction-tp4173961p4180628.html
Sent from the R help mailing list archive at Nabble.com.

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.
>
> ______________________________________________
> R-help at r-project.org 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