[R] regression

[William Dunlap]

The newdata argument to predict should be a data.frame (or environment or list)
containing the variables that are on the right side of the formula (the predictors).
In your case that means it should have a variable called 'Concentration'.
Since it didn't have such a variable (it contained only 'Replica.1'), predict found a
variable called 'Concentration' in the global environment and used it for the prediction.

To help avoid this problem, do not use the idiom
            Conc <- c(.1, .4, .9)
            Std <- c(2, 5, 7)
            myData <- data.frame(Conc, Std)
to make your data.frame: it leaves unwanted global variables around.
Instead use
            myData <- data.frame(
                 Conc = c(.1, .4, .9),
                 Std = c(2, 5, 7))
Then, if your newdata argument to predict is inappropriate you will get a
complaint that predict cannot find 'Conc'.

Bill Dunlap
Spotfire, TIBCO Software
wdunlap tibco.com


> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf
> Of Julen Tomás Cortazar
> Sent: Friday, September 13, 2013 6:16 AM
> To: r-help at r-project.org
> Subject: [R] regression
> 
> I am sorry,
> 
> 
> 
> I have a problem. When I use the "predict" function I am always obtaining the same
> result and I don't  know why. In adittion, the intercept and the residual values I get are
> wrong too.
> 
> 
> 
> std:
> 
> 
> 
> [1] 0.068 0.117 0.167 0.269 0.470 0.722
> 
> 
> 
> 
> 
> Concentration:
> 
> 
> 
> [1]   3.90625   7.81250  15.62500  31.25000  62.50000 125.00000
> 
> 
> 
> replica1:
> 
> 
> 
>   Replica.1
> 1     0.080
> 2     1.325
> 3     1.309
> 4     1.072
> 5     1.595
> 6     1.384
> 
> replica2:
> 
> Replica.2
> 1     0.098
> 2     1.335
> 3     1.271
> 4     1.187
> 5     1.569
> 6     1.268
> 
> 
> 
> 
> regresion <- lm(std ~ Concentration, mydata):
> 
> 
> 
> Call:
> lm(formula = std ~ Concentration, data = mydata)
> 
> Residuals:
>         1         2         3         4         5         6
> -0.035531 -0.007440  0.000742  0.019106  0.052834 -0.029711
> 
> Coefficients:
>                Estimate Std. Error t value Pr(>|t|)
> (Intercept)   0.0826219  0.0208118    3.97 0.016539 *
> Concentration 0.0053527  0.0003532   15.15 0.000111 ***
> ---
> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> 
> Residual standard error: 0.0366 on 4 degrees of freedom
> Multiple R-squared:  0.9829,   Adjusted R-squared:  0.9786
> F-statistic: 229.6 on 1 and 4 DF,  p-value: 0.0001106
> 
> 
> 
> 
> 
> > predict(regresion, replica1, int = "p")
>         fit          lwr       upr
> 1 0.1035309 -0.012098349 0.2191602
> 2 0.1244399  0.009958707 0.2389212
> 3 0.1662580  0.053716164 0.2787998
> 4 0.2498941  0.139724612 0.3600636
> 5 0.4171663  0.305409665 0.5289230
> 6 0.7517107  0.614489312 0.8889322
> 
> > predict(regresion, replica2, int = "p")
>         fit          lwr       upr
> 1 0.1035309 -0.012098349 0.2191602
> 2 0.1244399  0.009958707 0.2389212
> 3 0.1662580  0.053716164 0.2787998
> 4 0.2498941  0.139724612 0.3600636
> 5 0.4171663  0.305409665 0.5289230
> 6 0.7517107  0.614489312 0.8889322
> 
> 
> So, anyone knows what is happening to me?
> 
> Thank you very much!
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> This e-mail comes from CIC bioGUNE. The e-mail and any files transmitted with it are
> confidential and intended solely for the use of the individual or entity to whom they are
> addressed. Any unauthorised dissemination or copying of this e-mail or its attachments,
> and any use or disclosure of any information contained in them, is strictly prohibited and
> may be illegal. If you have received this e-mail in error, please notify or telephone + 34
> 944 06 13 00 and delete it from your system.
> 
> Please consider the environment before printing this email
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> 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.