Re: Confidence interval for all the estimated Y's?

Frank E Harrell Jr wrote:
Bill H wrote:
Jens wrote:

Dear all,

I'm regressing several random variables X with Y, total number of
observations is 1000++. I'm unsure how to calculate the confidence
interval for all the estimated Y's?

The residuals are "outlier free" (i.e. I iterated the regression
several times and removed such observations where the particular
residuals confidence interval [provided by the regression tool] didn't
cross the zero).

That will severely bias the resulting parameter estimates and especially
the variance.

Frank Harrell



Regards,
Jens


I'm worried when people so easily delete data. Did you check the
values to see if they were in error?

Bill H, Frank H

I tought it would be reasonable to remove such "outliers", there were
only less than 1% of such "outliers". Anyway I'll calculate the
confident interval with and without "outliers". The main question
remains open: how do I calculate the confidence interval for all the
estimated Y's?


Thanks,
Jens

.

Relevant Pages

  • Re: multivariate Cox model p values
    ... Frank E Harrell Jr wrote: ... parameter this translates to less bias in regression coefficients. ... variables together in one model (i.e. multivariable analysis), ...
    (sci.stat.consult)
  • Re: multivariate Cox model p values
    ... Frank E Harrell Jr wrote: ... no correlation with the outcome variable, ... one that produces a negative 'beta weight' --- a regression ... Frank Harrell ...
    (sci.stat.consult)
  • Re: distribution of an outlier?
    ... The idea behind such tagging is that if you generate ONE observation ... But in the analysis of residuals in a regression, the "outliers" are ... always the LARGEST observed residuals, ...
    (sci.stat.math)
  • Re: a method for detecting multivariate outliers..
    ... outlier detection,how would i come to know that clustering, regression ... Let me tell u all that i am not interested in doing clustering, ... Outliers can only be defined with respect to some model of ... detect outliers in case of multivariate data.. ...
    (sci.stat.consult)
  • Re: std. deviation
    ... particularly influential observations if they might be outliers. ... variables if there aren't many predictors. ... multiplicative rather than additive model; you don't say much about the ... Regression estimates aren't particularly sensitive to moderate outliers ...
    (sci.stat.consult)