multivariate linear regression

Hi,

I am working on a multivariate linear regression of the form y = Ax.

I am seeing a great dispersion of y w.r.t x. For example, the
correlations between y and x are very small, even after using some
typical transformations like log, power.

I tried with simple linear regression, robust regression and ace and
avas package in R (or splus). I didn't see an improvement in the fit
and predictions over simple linear regression. (I also tried this with
transformed variables)

I am sure that some of you came across such data. How did you deal with
it?

Linear regressions are good for the data like y = x +
0.01Normal(mu,sigma2) i.e. a small noise (data observed in a lab). But
linear regressions are bad for large noise, like typical market (or
survey) data.

Thank you,
Nagu

.

Relevant Pages

  • Re: test for outlier in time series
    ... Noise is constant, no dependance in noise. ... I get the idea of predicting an intervall with a certain confidence ... data and sudden level jumps neither. ... Which standard technique in linear regression are ...
    (sci.stat.math)
  • Re: test for outlier in time series
    ... Noise is constant, no dependance in noise. ... Concerning your question time series vs linear regression: ... data and sudden level jumps neither. ... Which standard technique in linear regression are ...
    (sci.stat.math)
  • eliminate noise from a regression model
    ... One of my collaegues is trying to plot a simple linear regression & ... finds that the r-sq is very low. ... eliminate the noise and replot the data. ...
    (sci.stat.edu)
  • Re: multivariate linear regression
    ... I am working on a multivariate linear regression of the form y = Ax. ... typical transformations like log, power. ... 0.01Normali.e. a small noise. ... Otherwise you are trying to fit noise. ...
    (sci.stat.consult)
  • Multivariate linear regression
    ... I am working on a multivariate linear regression of the form y = Ax. ... I am seeing a great dispersion of y w.r.t x. ... typical transformations like log, power. ... 0.01Normali.e. a small noise. ...
    (sci.stat.math)