[R] Problem with augPred?

Meissner, Tony (DFW) Tony.Meissner at sa.gov.au
Tue Sep 28 03:45:51 CEST 2010


I have a dataset, example of the data is shown below:

Grouped Data: drain_irr ~ irr | crop
        Year decades   crop irrisystem drain_irr    irr
1310 1995-96   1990s Citrus    various     0.400  9.021
1311 1995-96   1990s Citrus       drip     0.541  6.468
1312 1995-96   1990s Citrus   overhead     3.308 11.336
1313 1995-96   1990s Citrus  undertree     0.400  9.050
1314 1995-96   1990s Citrus  undertree     3.485 10.304
1315 1995-96   1990s Citrus  undertree     0.400  3.423
Grouped Data: drain_irr ~ irr | crop
        Year decades       crop irrisystem drain_irr  irr
3605 2002-03   2000s WineGrapes       drip   0.24738 5.89
3606 2002-03   2000s WineGrapes       drip   0.24738 5.89
3607 2002-03   2000s WineGrapes       drip   0.20230 5.95
3608 2002-03   2000s WineGrapes       drip   0.20230 5.95
3609 2002-03   2000s WineGrapes       drip   0.76890 6.99
3610 2002-03   2000s WineGrapes       drip   0.76890 6.99

I am using nlsList to fit a model  drain_irr~a0*irr^b0 for  the crop factor the output of which is shown below:
Call:
  Model: drain_irr ~ A0 * irr^B0 | crop
   Data: nswdat

Coefficients:
   A0
                            Estimate         Std. Error        t value              Pr(>|t|)
Other                0.00017782    0.00046336    0.38376           6.4108e-01
WineGrapes   0.00891031    0.00240466    3.70544           7.6315e-05
Citrus                 0.00142073    0.00092889    1.52949           3.1404e-02
DriedVine        0.03829533     0.01323868    2.89269           8.2945e-02

 B0
                             Estimate         Std. Error            t value           Pr(>|t|)
Other                3.9132              1.01438                3.8577           1.3867e-02
WineGrapes   2.5510             0.12029               21.2068           1.9062e-46
Citrus                3.0921              0.25447               12.1509           1.3566e-21
DriedVine       1.9547              0.14723                13.2762           4.3579e-09

Residual standard error: 0.53998 on 195 degrees of freedom

A plot of the data using qplot from the ggplot2 package gives the follow relationship




However, when I use augPred to eg augPred(nswdat.nls00) it gives the following graphic

While citrus and Other are OK compared with the qplot of the raw data, Winegrape and DriedVine are clearly not.  Have others encountered this problem with augPred?

I am using R-2.11.1 under Windows XP

Tschüß
Tony Meissner
Principal Scientist - Monitoring
Department for Water | Level 3 28 Vaughan Terrace Berri SA 5343
T 8595 2209  | M 0401 124 971
E tony.meissner at sa.gov.au<mailto:tony.meissner at sa.gov.au>

Mon | Tue | Wed | Thurs | Fri

www.waterforgood.sa.gov.au<http://www.waterforgood.sa.gov.au/> | www.sa.gov.au<http://www.sa.gov.au/>

The information in this e-mail may be confidential and/or legally privileged.
Use or disclosure of the information by anyone other than the intended recipient is prohibited and may be unlawful.
"Imagine" ©






More information about the R-help mailing list