Calculating The Joint Probability Of False Paternity Events

Moderator: MOD_nyhetsgrupper

Svar
D. Spencer Hines

Calculating The Joint Probability Of False Paternity Events

Legg inn av D. Spencer Hines » 04 aug 2007 19:47:31

I fear you are the one who is confused.

You obviously don't understand the mathematics of Joint Probabilities.

If the probability of an FPE is p = .02 [2%] in each generation of a descent
then the probability that the Paternal Descent is NOT false is p = .98.

Then calculating the JOINT Probability of True Paternal Descents over 50
generations gives us:

..98^50 = .3642.

Therefore the p for FPE over the 50 generations is:

1.0000 - .3642 = .6358 or about 64%.

However, using the 5% figure for FPE's we get a p of .9231 or 92%.

DSH

Lux et Veritas et Libertas

<j.s.plant@isc.keele.ac.uk> wrote in message
news:mailman.186.1186245752.31452.gen-medieval@rootsweb.com...

On Aug 2, 1:31 am, John Plant <j.s.pl...@isc.keele.ac.uk> wrote:
taf wrote:

Apart from a couple, whose male lines both originated in south
Lincolnshire around 1800, these 9 have random haplotypes. This is
consistent with expectation for a surname that originated from a single
ancestor, since, in the centuries since then, about half of the lines
are expected to have a false paternity event (FPE) somewhere in the line
of descent (unfaithful wife, adoption, unmarried mother giving child her
own surname, etc). This phenomenon of FPEs is widely documented in
DNA-genealogy literature.

The phenomenon has been widely discussed, but not widely documented.
To document it, you actually have to show that people who 'should'
have the same ancestor don't have the same type. All too often, though
this is assumed rather than documented. Are any of yours documented
FPEs?

taf

True. Let's see some documented FPE's as noted.

[...]

Perhaps you already knew all this; but most people as yet seem to get
confused.

John

Gjest

Re: Calculating The Joint Probability Of False Paternity Eve

Legg inn av Gjest » 05 aug 2007 00:51:54

Must break off now, time for bed in England.

Briefly, the experimental results in Table 1 of my Nomina 28 paper,
previously mentioned; and, more fully at:

http://www.plant-fhg.org.uk/dna.html

John

The 2% rate is probably far too low when it comes to the General Populace.

What "Experimental Results"?

You are siring potential FPE's?

Must be Great Fun...

DSH

j.s.plant@isc.keele.ac.uk> wrote in message
news:mailman.3.1186261190.7287.gen-medieval@rootsweb.com...

You're Welcome...

j.s.plant@isc.keele.ac.uk> wrote in message
news:mailman.1.1186257965.7287.gen-medieval@rootsweb.com...

DSH,

Thanks for the clarification. I do not claim to be immune from
confusion
myself. However, I was talking in terms of simple estimates and I do
not
consider the difference between 50% and 64% to be enormous when
compared
to other uncertainties in the estimates.

That all depends on whether you use the 2% or 5% rate.

With 5% we get an FPE rate of 92%

Following the meanderings you give us below and taking a 10% FPE rate
would give us a p [FPE] of .9948 over the 50 generations you postulate.

That's 99.48%.

A 30% FPE rate gives us a p[FPE] of 100% over the 50 generations.

DSH

Lux et Veritas et Libertas
---------------------------------------------------

I agree that the more precise calculations are essential for the higher
rates. However, I was restricting my estimates to the 2% rate because
that
was roughly what was relevant to my experimental results. The estimates
I
gave were adequate for the reasoning I presented.

John



-------------------------------
To unsubscribe from the list, please send an email to
GEN-MEDIEVAL-request@rootsweb.com with the word 'unsubscribe' without the
quotes in the subject and the body of the message

Gjest

Re: Calculating The Joint Probability Of False Paternity Eve

Legg inn av Gjest » 05 aug 2007 00:51:54

Must break off now, time for bed in England.

Briefly, the experimental results in Table 1 of my Nomina 28 paper,
previously mentioned; and, more fully at:

http://www.plant-fhg.org.uk/dna.html

John

The 2% rate is probably far too low when it comes to the General Populace.

What "Experimental Results"?

You are siring potential FPE's?

Must be Great Fun...

DSH

j.s.plant@isc.keele.ac.uk> wrote in message
news:mailman.3.1186261190.7287.gen-medieval@rootsweb.com...

You're Welcome...

j.s.plant@isc.keele.ac.uk> wrote in message
news:mailman.1.1186257965.7287.gen-medieval@rootsweb.com...

DSH,

Thanks for the clarification. I do not claim to be immune from
confusion
myself. However, I was talking in terms of simple estimates and I do
not
consider the difference between 50% and 64% to be enormous when
compared
to other uncertainties in the estimates.

That all depends on whether you use the 2% or 5% rate.

With 5% we get an FPE rate of 92%

Following the meanderings you give us below and taking a 10% FPE rate
would give us a p [FPE] of .9948 over the 50 generations you postulate.

That's 99.48%.

A 30% FPE rate gives us a p[FPE] of 100% over the 50 generations.

DSH

Lux et Veritas et Libertas
---------------------------------------------------

I agree that the more precise calculations are essential for the higher
rates. However, I was restricting my estimates to the 2% rate because
that
was roughly what was relevant to my experimental results. The estimates
I
gave were adequate for the reasoning I presented.

John



-------------------------------
To unsubscribe from the list, please send an email to
GEN-MEDIEVAL-request@rootsweb.com with the word 'unsubscribe' without the
quotes in the subject and the body of the message

Gjest

Re: Calculating The Joint Probability Of False Paternity Eve

Legg inn av Gjest » 05 aug 2007 00:51:54

Must break off now, time for bed in England.

Briefly, the experimental results in Table 1 of my Nomina 28 paper,
previously mentioned; and, more fully at:

http://www.plant-fhg.org.uk/dna.html

John

The 2% rate is probably far too low when it comes to the General Populace.

What "Experimental Results"?

You are siring potential FPE's?

Must be Great Fun...

DSH

j.s.plant@isc.keele.ac.uk> wrote in message
news:mailman.3.1186261190.7287.gen-medieval@rootsweb.com...

You're Welcome...

j.s.plant@isc.keele.ac.uk> wrote in message
news:mailman.1.1186257965.7287.gen-medieval@rootsweb.com...

DSH,

Thanks for the clarification. I do not claim to be immune from
confusion
myself. However, I was talking in terms of simple estimates and I do
not
consider the difference between 50% and 64% to be enormous when
compared
to other uncertainties in the estimates.

That all depends on whether you use the 2% or 5% rate.

With 5% we get an FPE rate of 92%

Following the meanderings you give us below and taking a 10% FPE rate
would give us a p [FPE] of .9948 over the 50 generations you postulate.

That's 99.48%.

A 30% FPE rate gives us a p[FPE] of 100% over the 50 generations.

DSH

Lux et Veritas et Libertas
---------------------------------------------------

I agree that the more precise calculations are essential for the higher
rates. However, I was restricting my estimates to the 2% rate because
that
was roughly what was relevant to my experimental results. The estimates
I
gave were adequate for the reasoning I presented.

John



-------------------------------
To unsubscribe from the list, please send an email to
GEN-MEDIEVAL-request@rootsweb.com with the word 'unsubscribe' without the
quotes in the subject and the body of the message

Gjest

Re: Calculating The Joint Probability Of False Paternity Eve

Legg inn av Gjest » 05 aug 2007 00:51:54

Must break off now, time for bed in England.

Briefly, the experimental results in Table 1 of my Nomina 28 paper,
previously mentioned; and, more fully at:

http://www.plant-fhg.org.uk/dna.html

John

The 2% rate is probably far too low when it comes to the General Populace.

What "Experimental Results"?

You are siring potential FPE's?

Must be Great Fun...

DSH

j.s.plant@isc.keele.ac.uk> wrote in message
news:mailman.3.1186261190.7287.gen-medieval@rootsweb.com...

You're Welcome...

j.s.plant@isc.keele.ac.uk> wrote in message
news:mailman.1.1186257965.7287.gen-medieval@rootsweb.com...

DSH,

Thanks for the clarification. I do not claim to be immune from
confusion
myself. However, I was talking in terms of simple estimates and I do
not
consider the difference between 50% and 64% to be enormous when
compared
to other uncertainties in the estimates.

That all depends on whether you use the 2% or 5% rate.

With 5% we get an FPE rate of 92%

Following the meanderings you give us below and taking a 10% FPE rate
would give us a p [FPE] of .9948 over the 50 generations you postulate.

That's 99.48%.

A 30% FPE rate gives us a p[FPE] of 100% over the 50 generations.

DSH

Lux et Veritas et Libertas
---------------------------------------------------

I agree that the more precise calculations are essential for the higher
rates. However, I was restricting my estimates to the 2% rate because
that
was roughly what was relevant to my experimental results. The estimates
I
gave were adequate for the reasoning I presented.

John



-------------------------------
To unsubscribe from the list, please send an email to
GEN-MEDIEVAL-request@rootsweb.com with the word 'unsubscribe' without the
quotes in the subject and the body of the message

Svar

Gå tilbake til «soc.genealogy.medieval»