If you are looking for robust multivariate or multivariable methods,
then review the Robust Methods Task View:
https://cran.r-project.org/view=Robust
--
David.
On 3/13/19 4:06 AM, Jackson, Daniel wrote:
Hi Frank and Dennis
I am in a similar situation but I would prefer to use a proportion
Thanks I am now in personal communication with Frank anyway!
Dan
-Original Message-
From: peter dalgaard
Sent: 14 March 2019 10:37
To: Jackson, Daniel
Cc: r-help@r-project.org
Subject: Re: [R] Rank ANCOVA
1. This is the R-help mailing list, not "Frank and Dennis"
2. You
1. This is the R-help mailing list, not "Frank and Dennis"
2. You seem to be referring to two posts from July 2010 by Frank Harrell and
Dennis Fisher.
-pd
> On 13 Mar 2019, at 12:06 , Jackson, Daniel
> wrote:
>
> Hi Frank and Dennis
>
> I am in a similar situation but I would prefer to use
Hi Frank and Dennis
I am in a similar situation but I would prefer to use a proportional odds
model. 2 questions.
1. Has rank ancova been implemented in R now, despite its short comings?
2. Where has it been shown to yield unreliable analyses? I would like this
evidence (which I believe
This has been shown to yield unreliable analyses. Use the more formal
proportional odds ordinal logistic model. This is a generalization of
the Wiloxon-Mann-Whitney-Kruskal-Wallis statistic. This is
implemented in the rms package and elsewhere.
Frank E Harrell Jr Professor and Chairman
Colleagues
I am trying to replicate some analyses performing in SAS using a "rank ANCOVA
model". I assume that this is a non-parametric ANOVA with covariates but I
have not been able to get confirmation from the people who did the original
analyses. Can anyone direct me to the comparable func
6 matches
Mail list logo
