Thank you for the clarification.
On Fri, Sep 11, 2020 at 2:36 PM Karl Schilling
wrote:
>
> Dear Luigi:
>
> no, µ1µ2 is not "missing"
>
> First, it should actually be "µ1 - µ2".
>
> And as your usual null-hypothesis when comparing h1 and h2 is that they
> are not different (i.e. µ1 = µ2), the latt
Dear Luigi:
no, µ1µ2 is not "missing"
First, it should actually be "µ1 - µ2".
And as your usual null-hypothesis when comparing h1 and h2 is that they
are not different (i.e. µ1 = µ2), the latter term adds up to 0 and may
be omitted.
Karl Schilling
__
Actually,
in the working example, Hutcheson himself did not report the term
µ1µ2: `t0 = h1-h2/(var1-var2)^1/2`. so I think we can live without it.
Case closed.
Thank you
On Fri, Sep 11, 2020 at 11:11 AM Luigi Marongiu
wrote:
>
> Hello,
> I have just realized in the original paper, the t test is d
Hello,
I have just realized in the original paper, the t test is defined as:
`t = h1-h2 -(µ1µ2)/(var1-var2)^1/2`. is the term -(µ1µ2) missing in
your formula? How to calculate µ1µ2?
Thank you
On Thu, Sep 10, 2020 at 2:41 PM Luigi Marongiu wrote:
>
> Update:
> I also added the confidence interval
Update:
I also added the confidence interval for the Shannon index:
```
#! Hutcheson's t-test for Shannon diversity equality
# thanks to Karl Schilling and Rui Barradas
hutcheson = function(A, B){
# compute Shannon index, variance and sum of elements
A_index <- Shannon(A)
B_index <- Shannon(B
Hello,
thank you for the code. To explain better, when I used vegan, I did
not count the species directly but simply prepared a dataframe where,
for each species, I counted the number of samples bearing such
species:
```
> str(new_df)
'data.frame': 3 obs. of 46 variables:
$ NC_001416 Enterobacte
Hello,
Sorry, there's an instruction missing. See inline.
Às 11:44 de 10/09/20, Rui Barradas escreveu:
If you want a function automating Karl's code, here it is. It returns an
object of S3 class "htest", R's standard for hypothesis tests functions.
The returned object can then be subset in the
If you want a function automating Karl's code, here it is. It returns an
object of S3 class "htest", R's standard for hypothesis tests functions.
The returned object can then be subset in the usual ways, ht$statistic,
ht$parameter, ht$p.value, etc.
library(QSutils)
hutcheson.test <- function
Update:
I can see that you used the function Shannon from the package QSutils.
This would supplement the iNext package I used and solve the problem.
Thank you.
On Thu, Sep 10, 2020 at 10:35 AM Luigi Marongiu
wrote:
>
> Thank you very much for the code, that was very helpful.
> I got the article b
Thank you very much for the code, that was very helpful.
I got the article by Hutcheson -- I don't know if I can distribute it
, given the possible copyrights, or if I can attach it here -- but it
does not report numbers directly: it refers to a previous article
counting bird death on a telegraph e
10 matches
Mail list logo
