Re: Appropriate journal for circular statistics paper
- From: J at KU <JPalikij@xxxxxxxxx>
- Date: Thu, 13 Mar 2008 14:59:44 -0700 (PDT)
On Feb 14, 9:38 am, rku...@xxxxxxxxxxxxxx (Rade Kutil) wrote:
I have written a paper and now I don't know where to send it. I am
not a pure mathematician and certainly not specialized on statistics.
Therefore, I am unsure about which journal is appropriate. Maybe
someone here can help me.
The paper has the title "Biased and Unbiased Estimation of the
Circular Mean Resultant Length and its Variance". It is about the
mean resultant length (MRL), which is the length (absolute value) of
the expected value of a complex random variable on the unit circle.
It serves as a measure of unimodal concentration of directional data.
In the past, only its estimator, i.e. mean over a sample of angles,
has been considered on its own in several different fields, e.g. as
PLI (phase locking index) on EEG data. But the properties of this
notion as an estimator have not been investigated.
My paper first considers the squared MRL, finds that it is biased,
develops an unbiased variant (similar to the unbiased empirical
variance), and conducts some numerical simulations to investigate the
bias of the non-squared MRL in both variants. Second, and estimator
of the estimator's variance and MSE is developed, both for the squared
and non-squared case. This is of practical importance when the
accuracy of an MRL estimation is to be judged. Again, numerical
simulations show the accuracy of the variance estimator. Finally, a
variance estimator for the MRL of a sample subset, based on the whole
sample, is developed.
While all this is everyday's business in the linear case
(e.g. variance of mean), the circular MRL case leads to quite
complicated and far from obvious results. It does not involve Hilbert
spaces, though, if you know what I mean.
Now, if you have read all this, you are already my dear friend. Does
some journal seem appropriate to you?
Rade
Best of luck Rade! Have you sent your paper off yet? I would suggest
taking an issue from each journal you are considering, and determining
which you simply like best. Look for length, diagrams, how the papers
are organized. Often times journals a specific in how papers are
arranged/written. Then, just send it off. The editor will be your
best friend/worst enemy. They might suggest changes, or different
journals.
Jason
.
- Prev by Date: Re: Using Ridge Regression to disentangle highly correlated explanatory variables
- Next by Date: SIGNS OF PROPHETHOOD
- Previous by thread: Using Ridge Regression to disentangle highly correlated explanatory variables
- Next by thread: SIGNS OF PROPHETHOOD
- Index(es):
Relevant Pages
|
