R help - Sep 2003 - estimating quantiles from binned data

Suppose I have a set of binned data, counts exceeding a series of
arbitrary thresholds, a total N, a minimum and maximum, those sorts of
things.  Is there a "standard" method for estimating arbitrary
quantiles from this?

My initial thought is that the counts and min/max give me solutions at
various points along the empirical cdf.  As the data are roughly
log-normal, I thought maybe I could use piece-wise log-normal
distributions between these points to estimate the arbitrary quantiles
I am interested in.

Are there "better thought out" methods than this?

Thanks!

-- 
Russell Senior         ``shtal latta wos ba padre u prett tu nashtonfi
seniorr at aracnet.com      mrlosh''  -- Bashgali Kafir for ``If you
have
                         had diarrhoea many days you will surely die.''

Have you considered making a normal probability plot?  This should help 
you evaluate the appropriatness of interpolating a log-normal.  The 
image of a mixture of lognormals would suggest limits on the accuracy of 
such interpolation.

hope this helps.
spencer graves

Russell Senior wrote:
> Suppose I have a set of binned data, counts exceeding a series of
> arbitrary thresholds, a total N, a minimum and maximum, those sorts of
> things.  Is there a "standard" method for estimating arbitrary
> quantiles from this?
> 
> My initial thought is that the counts and min/max give me solutions at
> various points along the empirical cdf.  As the data are roughly
> log-normal, I thought maybe I could use piece-wise log-normal
> distributions between these points to estimate the arbitrary quantiles
> I am interested in.
> 
> Are there "better thought out" methods than this?
> 
> Thanks!
>