divorce

"Dashi Toshii" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

"Scott Raymond" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

"Robert Chung" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

Scott Raymond wrote:

Quoted message said:

Quoted message said:

And in the same sense that we know the average lifespan of those that

died

Quoted message said:

last year (i.e. no projection is necessary), any projection that seeks

to

Quoted message said:

Quoted message said:

":guess" the lifespan of those that were born yesterday is using very

brown

Quoted message said:

numbers. How can we possibly anticipate the lifespan of those born last year when they will be
influenced by certain (but unknown) changes in medical care during the next 70 or so years?

Not too difficult to understand if you consider that the figures are

updated

Quoted message said:

at least annually.

Dashii

You lost me... we don't even know what will happen to Moore's Law (the law that says the number of
semiconductors placed on a given chip area will double every 18 months) in the next ten (let alone
seventy-some-odd) years. How can we know what will happen to life expectancies (an obviously much
more complex system)? I don't see how updating figures annually helps any.

Please help me understand.

Scott-

smiles said:

"Khoomei" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

You obviously haven't been through a divorce.

he was going through it before the tour ... and still won ... but had problems, it will help to
have it behind him ...

Yeah, I guess cancer may have also strengthened Lance to get it behind him. Man, Lance may become
some kind of super human being. Maybe Lance will win 10 Tours. Inspiring not only cancer
survivors everywhere, but divorce survivors everywhere. Maybe Lance will be cheered on by
divorced men at TdF 2004.

smiles said:

"Robert Chung" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

Well then, what proportion of marriages do you think will end in divorce or dissolution?

I will WASTE my time responding to you ... but will not WASTE it calculating a prediction on
divorce ...

So then how would you know that Kurg's reference was wrong?

Quoted message said:

BTW, it's 4 anyway ...

4% ??? Wow, you *are* a dumbass.

Scott Raymond said:

Quoted message said:

That projection, as well as all the projections I've seen since then, are based on synthetic
cohort calculations, not cross-sectional ones. Nuptiality and marital dissolution are central
topics in the study of family structure so researchers spend a lot of time calculating the
probability of marital dissolution. The technique is really no more complicated than the one used
for life tables, and insurance companies do that pretty routinely. Every projection I've seen
(and I'm betting I've seen a lot more of them than you) gives estimates of the probability that
first marriages in the US during the 1990's would end in divorce ranging from about 45% up to
about 60%.

I appreciate the fact that you've seen more projections than I have - and it really isn't hard to
be the case. I have yet to find a single attributable projection that has been published (vice
referenced by someone else). Admittedly, I haven't made a real effort to look for them (past the
internet), but if you have some studies in mind, I'd certainly appreciate the references.

As I wrote, these are very standard calculations. You can do a search on divorce projection stuff
by Cherlin or by Bumpass. I just googled up a paper that Bumpass and Martin wrote in 1989,
"Recent trends in marital disruption," Demography 26(1): 37-51. Look in the Journals: Demography,
Journal of Marriage and the Family, Population Studies, and the Sociology journals. The NCHS
stopped doing projections in the early 1990's for budgetary reasons and because academic
researchers were doing the projections anyway, but if for some reason you trust them more than
past-presidents of the Population Association of America you can find an old Current Population
Report P23 series from that era.

Quoted message said:

I'm curious about one aspect of your statement that "the probability that first marriages in the
US during the 1990's would end in divorce ranging from about 45% up to about 60%." How much of
total marriages do the 1990 marriages represent?

I don't quite understand your question. Are you asking how many first marriages that occurred in the
US in the year 1990 are still intact in 2003? BTW, part of that range I cited above depends on
whether "first marriage" includes marriages that are the first marriage for both partners or only
one of the partners (for example, first marriage for the husband but second marriage for the wife).
Some people count a first marriage by the bride's status.

Quoted message said:

And in the same sense that we know the average lifespan of those that died last year (i.e. no
projection is necessary), any projection that seeks to ":guess" the lifespan of those that were
born yesterday is using very brown numbers. How can we possibly anticipate the lifespan of those
born last year when they will be influenced by certain (but unknown) changes in medical care
during the next 70 or so years?

That's why demographers usually take a calculation of life expectancy at birth with a
(slightly)bigger grain of salt than they do a calculation of remaining life expectancy at age 75.
The technique is to construct a synthetic cohort. Demographers understand when estimates come from a
synthetic cohort calculation and when they come from true cohort calculations, so there's never any
confusion about how to interpret it. It's lay people who get confused. I can assure you that the
Social Security Administration has funded a lot of research on how to do better projections of life
expectancy. However, this turns out to be relevant to projections of marriage duration. Sadly, the
survivorship curve for marriages (some people say the decay rate or the hazard function, but I sort
of avoid that usage) is pretty steep so the projection isn't as squirrelly as you might have
thought. The life expectancy of a marriage is shorter than the life expectancy of a human so you're
"projecting" forward a shorter distance.

"Scott Raymond" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

"Dashi Toshii" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

"Scott Raymond" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

"Robert Chung" <[email hidden]> wrote in message
"]news:[email hidden]...

You lost me... we don't even know what will happen to Moore's Law (the

law

Quoted message said:

that says the number of semiconductors placed on a given chip area will double every 18 months) in
the next ten (let alone seventy-some-odd)

years.

Quoted message said:

How can we know what will happen to life expectancies (an obviously much more complex system)? I
don't see how updating figures annually helps

any.

Quoted message said:

Please help me understand.

OK.

A male baby born this year can expect to live for 76.3 years if he does not die sooner!

See, easy.

Dashii

"Robert Chung" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

As I wrote, these are very standard calculations. You can do a search on divorce projection stuff
by Cherlin or by Bumpass. I just googled up a paper that Bumpass and Martin wrote in 1989, "Recent
trends in marital disruption," Demography 26(1): 37-51. Look in the Journals: Demography, Journal
of Marriage and the Family, Population Studies, and the Sociology journals. The NCHS stopped doing
projections in the early 1990's for budgetary reasons and because academic researchers were doing
the projections anyway, but if for some reason you trust them more than past-presidents of the
Population Association of America you can find an old Current Population Report P23 series from
that era.

Thanks for the two names (Cherlin and Bumpass)... they've pointed me towards some very interesting
reading. Especially one publication by Bumpass that looks at the divorce rate among age (when
married) groups, education, and race. I haven't looked at it in depth, but I'm sure it will indicate
that white, educated, and late-marriers have a lower divorce rate than black, uneducated, and
early-marriers. To bring this back on topic, I'll be curious to learn where Lance fits in.

Quoted message said:

Quoted message said:

I'm curious about one aspect of your statement that "the probability that first marriages in the
US during the 1990's would end in divorce ranging from about 45% up to about 60%." How much of
total marriages do the 1990 marriages represent?

I don't quite understand your question. Are you asking how many first marriages that occurred in
the US in the year 1990 are still intact in 2003? BTW, part of that range I cited above depends on
whether "first marriage" includes marriages that are the first marriage for both partners or only
one of the partners (for example, first marriage for the husband but second marriage for the
wife). Some people count a first marriage by the bride's status.

I was asking what percentage of total marriages in existance today do the 1990's (1990-1999)
marriages represent.

Quoted message said:

Quoted message said:

And in the same sense that we know the average lifespan of those that died last year (i.e. no
projection is necessary), any projection that seeks to ":guess" the lifespan of those that were
born yesterday is using very brown numbers. How can we possibly anticipate the lifespan of those
born last year when they will be influenced by certain (but unknown) changes in medical care
during the next 70 or so years?

That's why demographers usually take a calculation of life expectancy at birth with a
(slightly)bigger grain of salt than they do a calculation of remaining life expectancy at age 75.
The technique is to construct a synthetic cohort. Demographers understand when estimates come from
a synthetic cohort calculation and when they come from true cohort calculations, so there's never
any confusion about how to interpret it.

My point was that the environment (which affects both the life expectancy and the marriage
expectancy) may very well change in the future. This is more likely in the "far" future of death
(70+ years from birth) than the "near" future of divorce - as you've stated below.

Quoted message said:

It's lay people who get confused. I can assure you that the Social Security Administration has
funded a lot of research on how to do better projections of life expectancy. However, this turns
out to be relevant to projections of marriage duration. Sadly, the survivorship curve for
marriages (some people say the decay rate or the hazard function, but I sort of avoid that usage)
is pretty steep so the projection isn't as squirrelly as you might have thought. The life
expectancy of a marriage is shorter than the life expectancy of a human so you're "projecting"
forward a shorter distance.

I'd rather see the hazard function - it would give a much better picture of the dynamics of divorce.
I'm certain that marriages past a duration of 40 years make up a very small percentage of the total
divorces. Why do you avoid the hazard function?

OK, again, thanks for the references. I appreciated it.

Scott-

Scott Raymond said:

I was asking what percentage of total marriages in existance today do the 1990's (1990-1999)
marriages represent.

I don't know. That's not a question that demographers usually ask because it depends both on the
size of the marriage cohorts in a given year and the width of the time interval, so it varies too
much to tell you anything systematic. We're usually more interested in the rates or, as I indicated,
the hazard function, which is an instantaneous rate.

Quoted message said:

My point was that the environment (which affects both the life

expectancy

Quoted message said:

and the marriage expectancy) may very well change in the future

That's exactly right, but the people who use these things (insurance companies, pension funds, the
Social Security Administration, health care planners, etc) understand the limitations.

Quoted message said:

I'd rather see the hazard function - it would give a much better picture of the dynamics of
divorce. I'm certain that marriages past a duration of 40 years make up a very small percentage of
the total divorces. Why do you avoid the hazard function?

I don't avoid the hazard function, I just avoid the use of the term when talking about marital
dissolution. Some people think it pejorative and they then focus on the word and not the concept.

Dashi Toshii said:

A male baby born this year can expect to live for 76.3 years if he does not die sooner!

That is incorrect. Anyone who does not die before age 76.3 either will die at that age, or later.
There are who will die in that instant, others will live longer. So on average, all who will not
die before
76.3, can expect to live longer than that.

Thanks,
E.

GregR <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

Robert Chung said:

Jeff Jones said:

It's official: cyclingnews.comnewsOpen ↗

Armstrong: "The craziest thing is, we're closer now and better friends than ever before."

That's [censored]. Either he is afraid the divorce may tarnish his image (which means lower
endorsement $$) or a divorce tactic. Lull them into a false sense of security, then when the
proceedings commence it gets ugly.

G

Even if LA was cheating on Kik, it still is a bit unreasonable for Kik to want out of the
marriage. She is just a "Trophy Wife", who is supposed to find gratification in vicariously
enjoying the successes of her husband. Does she have any real talent for anything? She seems to
want to embark on some journey of self-actualization, but one has to wonder how much "self" is
there to actualize. Frankly, Lance's career is more important than she is, and she ought to have
abided the marriage for another 3-4 years, until LA was well and truly retired. If she needed to
take a lot of Valium to make it through, and have a Nanny raise the kids, so be it.

She chose to marry a celebrity sports hero, with all that entails. There is a price to pay for
the Lifestyle that comes along with the millions that Armstrong earns. There really is no
justification for her wanting to get a divorce at this stage, unless Lance beats her, or is
subjecting her to great mental cruelty and abuse. So far, there have been no accusations of
that. Just marital infidelity.

Fortunately, this is all going down nearly a year before the 2004 TdF, so that there is a good
chance that LA will be able to properly focus on winning his 6th TdF. I don't believe that the
divorce will hurt LA's cycling performance. Hopefully, July 2004 won't be as hot in France as
it was this year

"Robert Chung" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

Scott Raymond said:

I was asking what percentage of total marriages in existance today do the 1990's (1990-1999)
marriages represent.

I don't know. That's not a question that demographers usually ask because it depends both on the
size of the marriage cohorts in a given year and the width of the time interval, so it varies too
much to tell you anything systematic. We're usually more interested in the rates or, as I
indicated, the hazard function, which is an instantaneous rate.

I'm not sure who the "we" are, but I'd think most of those that get married are interested in their
relative risk to divorce at marriage *AND* the shape of the hazard function. My background isn't in
demographics - rather I'm an engineer - but "reliability" seems a common concept in both. What is
the reliability that a system (propulsion, generator, marriage, ...) will fail given it has already
survived X days/years?

I have one more favor to ask - and it gets to what was supposed to be my original point: Do you know
of any studies that compare life expectancy calcualted at birth (say from 1900) to actual average
life (most people born in 1900 are already dead!)? I'd think these two figures are markedly
different. How much certainy is there in the predicition of the divorce rate for marriages in 2003?

Wow... this is certainly *way* off topic.

Scott-

"Isidor Gunsberg" <[email hidden]> wrote in message

Quoted message said:

Even if LA was cheating on Kik, it still is a bit unreasonable for Kik to want out of the
marriage. She is just a "Trophy Wife", who is supposed to find gratification in vicariously
enjoying the successes of her husband. Does she have any real talent for anything? She seems
to want to embark on some journey of self-actualization, but one has to wonder how much "self"
is there to actualize. Frankly, Lance's career is more important than she is, and she ought to
have abided the marriage for another 3-4 years, until LA was well and truly retired. If she
needed to take a lot of Valium to make it through, and have a Nanny raise the kids, so be it.

She chose to marry a celebrity sports hero, with all that entails. There is a price to pay for
the Lifestyle that comes along with the millions that Armstrong earns. There really is no
justification for her wanting to get a divorce at this stage, unless Lance beats her, or is
subjecting her to great mental cruelty and abuse. So far, there have been no accusations of
that. Just marital infidelity.

Fortunately, this is all going down nearly a year before the 2004 TdF, so that there is a
good chance that LA will be able to properly focus on winning his 6th TdF. I don't believe
that the divorce will hurt LA's cycling performance. Hopefully, July 2004 won't be as hot in
France as it was this year

That's all good and well except AFAIK we don't know which one of them filed for divorce.

Scott Raymond said:

I have one more favor to ask - and it gets to what was supposed to be my original point: Do you
know of any studies that compare life expectancy calcualted at birth (say from 1900) to actual
average life (most people born in 1900 are already dead!)? I'd think these two figures are
markedly different. How much certainy is there in the predicition of the divorce rate for
marriages in 2003?

I think "life expectancy" should be instead called "birth expectancy", as it instead calculated the
expected birth date of people who die in a given year. In the "media", it's often mislabeled as the
expected life span of someone born at a given moment, which is nonsense, of course.

A real "life expectency" calculation would include extrapolation and its associated increased
uncertainties, ending up being largely meaningless.

Clarification -- the statistic has to be birth-rate normalized, of course.

So, if the probability distribution of someone being born at time t' and dying at time t is p(t',t),
and the birth rate at time t' is r(t'😉, then one calculates an effective statistic mislabeled as
"life expectency", tau(t) :

/ t
|
| dt' t' p(t', t) / r(t'😉
|
/-infy
tau(t) = t - --------------------------
/ t
|
| dt' p(t', t)/ r(t'😉
|
/ -infy

However, this isn't the actual life expectency at time t, which would be :

/ infy
|
| dt' t' p(t, t'😉
|
/ t tau'(t) = -------------------- - t / infy
|
| dt' p(t, t'😉
|
/ t

Daniel Connelly said:

I think "life expectancy" should be instead called "birth expectancy", as it instead calculated
the expected birth date of people who die in a given year. In the "media", it's often mislabeled
as the expected life span of someone born at a given moment, which is nonsense, of course.

A real "life expectency" calculation would include extrapolation and its associated increased
uncertainties, ending up being largely meaningless.

The statistic has nothing to do with someone born this year.

See my other post.

WRT divorce, the divorce rate in the US is approximately 1 per second (
nationmaster.comPeopleOpen ↗ for 1990 data ) This is approximately half the
marriage rate. If the marriage rate is an increasing function of time over the past 20 years or so,
this means most marriages (although perhaps not most first marriages) have ended in divorce. So
Armstrong and Kristin divorcing isn't unusual.

Dan

Ewoud Dronkert said:

Dashi Toshii said:

A male baby born this year can expect to live for 76.3 years if he does not die sooner!

That is incorrect. Anyone who does not die before age 76.3 either will die at that age, or later.
There are who will die in that instant, others will live longer. So on average, all who will not
die before
76.3, can expect to live longer than that.

Thanks,
E.

"Daniel Connelly" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

Scott Raymond said:

I have one more favor to ask - and it gets to what was supposed to be my original point: Do you
know of any studies that compare life expectancy calcualted at birth (say from 1900) to actual
average life (most people

born

Quoted message said:

Quoted message said:

in 1900 are already dead!)? I'd think these two figures are markedly different. How much
certainy is there in the predicition of the divorce rate for marriages in 2003?

I think "life expectancy" should be instead called "birth expectancy", as

it

Quoted message said:

instead calculated the expected birth date of people who die in a given

year.

Quoted message said:

In the "media", it's often mislabeled as the expected life span of someone born at a given moment,
which is nonsense, of course.

Yup... that's my point. Looking backward is (relatively) easy. Looking forward is difficult, mostly
because the model is likely to change over such a long period. But I'm still curious to see a
comparison between a forward-looking model and a backward-looking data set. At first glanse, I
thought the table at cdc.gov02hus028.pdfOpen ↗ might be of use - but
it seems to be all backward looking.

Scott-

Daniel Connelly said:

The statistic has nothing to do with someone born this year.

I guess in the sense that the failure rate for light bulbs from a particular manufacturing line has
nothing to do with light bulbs manufactured today.

Quoted message said:

WRT divorce, the divorce rate in the US is approximately 1 per second (
nationmaster.comPeopleOpen ↗ for 1990 data ) This is approximately half the
marriage rate. If the marriage rate is an increasing function of time over the past 20 years or
so, this means most marriages (although perhaps not most first marriages) have ended in divorce.
So Armstrong and Kristin divorcing isn't unusual.

Dan

This would be the case for what is (mis-) named a "stable population"
(i.e., where the age structure is an eigenvector for a matrix of birth and death rates. In that case
the eigenvalue for that matrix is the equilibrium growth rate for the population) with fixed
marriage and divorce functions. On the one hand, the U.S. population is far from stable, so
that estimate you're making will be off; on the other, the divorce function (the hazard rate)
is steep enough that the non-stability of the population isn't *that* important because the
divorces tend to lag the marriages by not-so-much time. That's why the ratio of the divorce
rate to the marriage rate may not be right, but it is in the right ballpark.

"Daniel Connelly" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

The statistic has nothing to do with someone born this year.

See my other post.

WRT divorce, the divorce rate in the US is approximately 1 per second (
nationmaster.comPeopleOpen ↗ for 1990 data ) This is approximately half the
marriage rate.

That sounds suspiciously like 50%.

Daniel Connelly said:

Clarification -- the statistic has to be birth-rate normalized, of course.

So, if the probability distribution of someone being born at time t' and dying at time t is
p(t',t), and the birth rate at time t' is r(t'😉, then one calculates an effective statistic
mislabeled as "life expectency", tau(t) :

/ t
|
| dt' t' p(t', t) / r(t'😉
|
/-infy
tau(t) = t - --------------------------
/ t
|
| dt' p(t', t)/ r(t'😉
|
/ -infy

Uh, Dan, this is not what life expectancy is.

Quoted message said:

However, this isn't the actual life expectency at time t, which would be :

/ infy
|
| dt' t' p(t, t'😉
|
/ t tau'(t) = -------------------- - t / infy
|
| dt' p(t, t'😉
|
/ t

Um, this isn't it, either.

However, you're right that there are two different life expectancies: one calculated for a birth
cohort (and can only be calculated after the entire cohort has died) and the period life expectancy
(which is usually what you see printed in the newspapers), which uses the probabilities of death in
the year t for people age a (i.e., born in year t-a). If you think of a surface where one axis is
age, one axis is time, and the height of the surface is the proportion surviving of each birth
cohort, the cohort expectation of life is the integral along the 45 degree diagonal while the period
expectation of life is the integral along the time axis.

"Robert Chung" <[email hidden]> wrote in message "]news:[email hidden]...

Quoted message said:

Quoted message said:

I will WASTE my time responding to you ... but will not WASTE it calculating a prediction on
divorce ...

So then how would you know that Kurg's reference was wrong?

"projections" are not based on fact ... just theortical numbers based on trends ... if i wanted I
could bend any "projection" i wanted based on the sample ... don't be ignorant all your life.

Quoted message said:

Quoted message said:

BTW, it's 4 anyway ...

4% ??? Wow, you *are* a dumbass.

and you are not?? if you were not one you would have known the movie reference;-) Did you see a
% sign ...

s boardnbike.comboardnbike.comOpen ↗

Robert Chung:

Quoted message said:

Uh, Dan, this is not what life expectancy is.

What's the formula, then?

Quoted message said:

Um, this isn't it, either.

However, you're right that there are two different life expectancies: one calculated for a birth
cohort (and can only be calculated after the entire cohort has died) and the period life
expectancy (which is usually what you see printed in the newspapers), which uses the probabilities
of death in the year t for people age a (i.e., born in year t-a). If you think of a surface where
one axis is age, one axis is time, and the height of the surface is the proportion surviving of
each birth cohort, the cohort expectation of life is the integral along the 45 degree diagonal
while the period expectation of life is the integral along the time axis.

Right. But neither is the "expectation value of the time to death of a baby born now", which is
commonly reported.

Dan