Re: [R] linear models with factors

2011-02-12 Thread Bert Gunter
Ummm... The simple answer to your question: i want the estimates of coefficients of each level along with the intercept term(general effect).can anyone please help me? is no, because it's matrhematically impossible. You need to consult a local statistician or read up on linear mo

Re: [R] linear models with factors

2011-02-12 Thread David Winsemius
On Feb 12, 2011, at 11:16 AM, Ista Zahn wrote: Well, I may have been too hasty there. Actually the question was slightly improved. I think you are looking for ?contr.sum Another issue will arise besides what the switch of the contrasts -- The intercept is not a "general effect" but rather a

Re: [R] linear models with factors

2011-02-12 Thread Christophe Pallier
See ?contrasts and ?contr.sum By default, R uses contr.treatment, you want contr.sum Christophe On Sat, Feb 12, 2011 at 6:57 AM, ATANU wrote: > > i am trying to fit a linear model with both continuous covariates and > factors. When fitted with the intercept > term the first level of the factor

Re: [R] linear models with factors

2011-02-12 Thread Ista Zahn
Well, I may have been too hasty there. Actually the question was slightly improved. I think you are looking for ?contr.sum Best, Ista On Sat, Feb 12, 2011 at 3:55 PM, Ista Zahn wrote: > I already responded to your identical post yesterday, asking you to > actually ask a question. "can you please

Re: [R] linear models with factors

2011-02-12 Thread Ista Zahn
I already responded to your identical post yesterday, asking you to actually ask a question. "can you please help me?" is not enough. Best, Ista On Sat, Feb 12, 2011 at 5:57 AM, ATANU wrote: > > i am trying to fit a linear model with both continuous covariates and > factors. When fitted with th

[R] linear models with factors

2011-02-12 Thread ATANU
i am trying to fit a linear model with both continuous covariates and factors. When fitted with the intercept term the first level of the factor is treated by R as intercept and the estimate of the effects of remaining levels(say i th level) are given as true estimate of i th level - estimate o

Re: [R] linear models with factors

2011-02-11 Thread Ista Zahn
I don't see any question here, other than "can you please help me". Since long-awaited esp package has still not been released, you're going to have to be more specific than that... Best, Ista On Fri, Feb 11, 2011 at 2:05 PM, ATANU wrote: > > i am trying to fit a linear model with both continuou

[R] linear models with factors

2011-02-11 Thread ATANU
i am trying to fit a linear model with both continuous covariates and factors. When fitted with the intercept term the first level of the factor is treated by R as intercept and the estimate of the effects of remaining levels(say i th level) are given as true estimate of i th level - estimate o