[R] Re gression function lm() not giving proper results

[Alain Zuur]

Read the warning message! It has converted your variables into factors.
Figure out why...and you will have solved the problem.


Alain Zuur



moumita wrote:
> 
> *
> *
> 
> Hi ,
> 
> Can anyone help me please  with this problem?*
> *
> 
> *CASE-I*
> 
> all_raw_data_NAomitted is my data frame.It has columns with names i1 ,i2,
> i3,i4
, till i15.It has 291 rows actually ,couldn’t show here.
> 
> The data frame looks like this:--
> 
>        i1 i2 i3 i4 i5 i6 i7 i8 i9 i10 i11 i12 i13 i14 i15
> 
> 2    2  2  2  2  2  2  2  2  2   2   1   2   2     3           2
> 
> 3    2  2  2  2  3  2  2  3  3   3   2   3   3    3            3
> 
> 4    2  2  2  2  2  2  2  1  1   1   2   1   2    2            2
> 
> 6    2  2  1  2  1  1  2  2  1   1   1   1   2    2           2
> 
> 8    3  2  2  2  3  3  3  2  3   2   3   2   3    3            2
> 
> 9    2  2  2  2  2  2  3  3  3   2   3   3   3    2           2
> 
> 10   1  1  1  1  1  1  1  1  1   1   1   1   1    1          1
> 
> 12   2  2  2  3  2  2  2  1  3   2   1   2   2    3           3
> 
> 
> 
> While doing regression  i1 being the dependent variable and i2 as the
> predictor  the outputs produced are not correct.The o/ps are as shown
> below:---
> 
> *all_raw_data_NAomitted$i1<-as.vector(as.matrix(all_raw_data_NAomitted$i1))
> all_raw_data_NAomitted$i2<-as.vector(as.matrix(all_raw_data_NAomitted$i2))
> *
> 
> *
> *
> 
> *fit<-lrm(i1 ~ i2 + NULL,all_raw_data_NAomitted)*
> 
>> source("regression.R")
> 
> [1] "Printing regression value........................."
> 
> Call:
> 
> lm(formula = i1 ~ i2, data = all_raw_data_NAomitted)
> 
> Residuals:
> 
>      Min       1Q   Median       3Q      Max
> 
> -1.46154 -0.19277 -0.03529 -0.03529  1.96471
> 
> *Coefficients:*
> 
> *            Estimate Std. Error t value Pr(>|t|)*
> 
> *(Intercept)  1.19277    0.05302   22.50   <2e-16 ****
> 
> *i22          0.84252    0.06469   13.03   <2e-16 ****
> 
> *i23          1.52723    0.11021   13.86   <2e-16 ****
> 
> *i24          2.26877    0.14409   15.74   <2e-16 ****
> 
> ---
> 
> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> 
> 
> 
> Residual standard error: 0.4831 on 287 degrees of freedom
> 
> Multiple R-squared: 0.5815,     Adjusted R-squared: 0.5771
> 
> F-statistic: 132.9 on 3 and 287 DF,  p-value: < 2.2e-16
> 
> 
> 
> Error in main() :
> 
> In addition: Warning messages:
> 
> 1: In model.matrix.default(mt, mf, contrasts) :
> 
>   variable 'i1' converted to a factor
> 
> 2: In model.matrix.default(mt, mf, contrasts) :
> 
>   variable 'i2' converted to a factor
> 
> *The results produced are incorrect and do not match with SPSS results
> ,you
> can find it out having a look at the coefficients sections of the
> result.my
> variables were i1 and i2.*
> 
> 
> 
> *CASE-II*
> 
> Whereas  if I do this the results produced are correct:--
> 
>> d1<-c(1,2,3,NA,6,7,8)
> 
>> d2<-c(2,3,4,3,1,2,2)
> 
>> d3<-c(2,1,2,1,2,1,3)
> 
>> d4<-c(5,6,2,1,1,2,2)
> 
>> d<-data.frame(d1,d2,d3,d4)
> 
>> d
> 
>   d1 d2 d3 d4
> 
> 1  1  2  2  5
> 
> 2  2  3  1  6
> 
> 3  3  4  2  2
> 
> 4 NA  3  1  1
> 
> 5  6  1  2  1
> 
> 6  7  2  1  2
> 
> 7  8  2  3  2
> 
>> fit<-lm(d1 ~ d2+d3+d4)
> 
>> summary(fit)
> 
> 
> 
> Call:
> 
> lm(formula = d1 ~ d2 + d3 + d4)
> 
> 
> 
> Residuals:
> 
>       1       2       3       5       6       7
> 
> -1.7865  0.9698 -1.2250 -1.4802  1.2761  2.2459
> 
> 
> 
> Coefficients:
> 
>             Estimate Std. Error t value Pr(>|t|)
> 
> (Intercept)   9.1912     5.1807   1.774    0.218
> 
> d2           -0.7570     1.2208  -0.620    0.598
> 
> d3            0.0151     1.7474   0.009    0.994
> 
> d4           -0.9842     0.6772  -1.453    0.283
> 
> 
> 
> Residual standard error: 2.692 on 2 degrees of freedom
> 
>   (1 observation deleted due to missingness)
> 
> Multiple R-squared: 0.6507,     Adjusted R-squared: 0.1267
> 
> F-statistic: 1.242 on 3 and 2 DF,  p-value: 0.4751
> 
> In case – (I) if I make the individual columns as vectors also ,I do not
> get
> correct results.what could be the cause of the incorrect results produced.
> 
> 
> -- 
> Thanks
> Moumita
> 
> 	[[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.
> 
> 

-- 
View this message in context: http://www.nabble.com/Regression-function-lm%28%29-not-giving-proper-results-tp24569513p24569838.html
Sent from the R help mailing list archive at Nabble.com.