[R] slope coefficient of a quadratic regression bootstrap

[Rui Barradas]

Hello,

You probably mean

fit2 <-lm(logDIS~logBMN+I(logBMN^2), data=d)  # d, not dataN


Hope this helps,

Rui Barradas
Em 06-05-2013 00:51, Elaine Kuo escreveu:
> Hello all,
>
> I found the following code worked.
>
> code
> require(boot)
> function.fit2  <-  function(data,i){
>            d  <-  data[i,]   #  select  obs.  in  bootstrap  sample
>            fit2 <-lm(logDIS~logBMN+I(logBMN^2), data=dataN)
>            coefficients(fit2)   #  return  coefficient  vector
>            }
>
> boot.fit2  <-  boot(dataN,  function.fit2,  1000)
>
>
>
> On Mon, May 6, 2013 at 7:36 AM, Elaine Kuo <elaine.kuo.tw at gmail.com> wrote:
>
>> Hello,
>>
>> I want to know if two quadratic regressions are significantly different.
>> I was advised to make the test using
>> step 1 bootstrapping both quadratic regressions and get their slope
>> coefficients.
>>            (Let's call the slope coefficient *â*^1 and *â*^2)
>> step 2 use the slope difference *â*^1-*â*^2 and bootstrap the slope
>> coefficent
>> step 3   find out the sampling distribution above and calculate the % =0
>> step 4  multiple the % by 2
>>
>> However, I am new to the package boot.
>> I wrote a code for step 1 and got a error message.
>>
>> Please kindly advise how to modify it and thank you.
>>
>> Elaine
>>
>> code
>> require(boot)
>> function.fit2  <-  function(data){
>>            #data  <-  data[indices,]   #  select  obs.  in  bootstrap
>>   sample
>>            fit2 <-lm(logDIS~logBMN+I(logBMN^2))
>>            coefficients(fit2)   #  return  coefficient  vector
>>            }
>>
>> boot.fit2  <-  boot(dataN,  function.fit2,  1000)
>>
>> error
>>> fit2.boot  <-  boot(dataN,  fun.fit2,  1000)
>> Error in statistic(data, original, ...) : unused argument (original)
>>
>>
>
> 	[[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.
>