[Forensim-help] Fwd: Fwd: Forensim-help post from phil.danielson at du.edu requires approval

[hinda haned]

Hi Keith,
My answer concerned dropout, for drop-in, it was based on our 'worst case
scenario' so it is indeed very high.
Thanks
Hinda


2015-06-25 17:34 GMT+02:00 Keith Inman <keith.inman at csueastbay.edu>:

> The two or three labs (with which I'm familiar)  that have derived their
> own dropin probabilities empirically from data have found it to be much
> less than .01.
>
> Is it possible, Hinda, that your local sources were including things like
> artifacts (dye blobs, high stutter) in their estimation? That would
> certainly drive up a P(Di).
>
> Best......Keith
>
> On Thu, Jun 25, 2015 at 7:17 AM, hinda haned <hindantation at gmail.com>
> wrote:
>
>> Hi Phil,
>> I relied on expert prior belief, I asked reporting officers what they
>> considered as 'moderate dropout' for an NGM profile.
>> This default value, while useful, is of course subjective, and an
>> estimate based on available data is preferred.
>> Hope this helps,
>> Hinda
>>
>>
>>
>> ---------- Message transféré ----------
>> From: Phil Danielson <phil.danielson at du.edu>
>> To: "forensim-help at lists.r-forge.r-project.org" <
>> forensim-help at lists.r-forge.r-project.org>
>> Cc:
>> Date: Wed, 24 Jun 2015 21:23:40 +0000
>> Subject: Drop Out Probability
>>
>> How was the default dropout probability of 0.1 determined for LRmix?
>>
>>
>>
>>
>> ---------- Message transféré ----------
>> From: forensim-help-request at lists.r-forge.r-project.org
>> To:
>> Cc:
>> Date: Wed, 24 Jun 2015 23:23:53 +0200
>> Subject: confirm 04a7daf8ee926f115bac7de0a6b63d713a8f9dca
>> If you reply to this message, keeping the Subject: header intact,
>> Mailman will discard the held message.  Do this if the message is
>> spam.  If you reply to this message and include an Approved: header
>> with the list password in it, the message will be approved for posting
>> to the list.  The Approved: header can also appear in the first line
>> of the body of the reply.
>>
>>
>>
>> --
>> Hinda
>>
>> _______________________________________________
>> Forensim-help mailing list
>> Forensim-help at lists.r-forge.r-project.org
>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/forensim-help
>>
>
>
>
> --
> Keith Inman
> Associate Professor
> Criminal Justice Administration
> California State University East Bay
> 25800 Carlos Bee Blvd
> Hayward, CA 94542
> keith.inman at csueastbay.edu
>



-- 
Hinda



-- 
Hinda
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/forensim-help/attachments/20150702/a264c538/attachment.html>