Re message 68 from Mr.Edward MENSAH – with all due respect to this
person, group members are strongly advised NOT to co-operate with any
requests of this nature.  I and many other people get large numbers
of such emails.  The great majority of them are understood to be
either money-laundering exercises or scams in which the respondent is
duped to part with his money for nothing.  So, the message is, don't
touch them with a bargepole.  (Also, don't open any attachment or go
to any website that an email points you to, unless you're sure it's
trustworthy – there are lots of viruses around, and those sources are
the highest risk for transmission).

Group instigator – the onus is on you to sort out how to prevent
material of this kind getting in – I don't know how it's done, but
other list owners manage it.


--- In Statisticians_group@..., "alma_dezor"
<alma_dezor@y...> wrote:
> Reply to alma_muse1@w...
>
> l am Mr.Edward MENSAH from Burkina-faso,l represent the local
miners
> because of my family's status in my province. We have in stock big
> quantity of gold dust for sale, in benin republic .
> Because of high cost of export and shipment procedure couple with
my
> government policies to monopolize gold business. Based on this
> procedure you are advice to come to benin republic cotonou for a
> round table business discussion and understanding then a working
> agreement will be signed by both parties,the gold dust will be
> presented for inspection as we don't do gold business in absent.
> During the meetting in benin republic we will propose to you the
> best option to conclude a safe and risk free gold transaction for
> the security of the buyer,the sellers money and machandise.
> We assure you good business relationship,safe delivery of your
goods
> to any refinery of your choice.
> The specification of the product is as follows:
>
> i)   Product: Alluvial gold dust (A U METAL)
> ii)  Quantity:500kgs,250kgs also+ in Burkina_faso.
> iii)  Quality:22 + + Carats.
> iv)  Moisture:0.25%.
> v)   Purity:92.%.
> vi)  Price:usd 5000.per kg
> vii) Location:cotonou republic du Benin or Burkina faso An
> accridated custom agent will be introduced by us to arrange all the
> necessary export papers to avoid any problem at the airport during
> shipment of the gold to any refinery of your choice.
>
> Best regards
> EDWARD MENSAH
> Reply to alma_muse1@w...

Here you will get parameter estimates along with 95% confidence interval for parameter estimates.

Hallo everybody,

I am looking for computer algorithms, preferably in Excel VBA, for
calculating non-linear regression parameters. I am particularly
interested in the following non-linear regression models:

- Logarithmic model: y = a + b * ln(x)
- Exponential model: y = a * e ^ (b * x)
- Power model:       y = a * x ^ b
- Inverse model:     y = a + b / x
- Quadratic model:   y = a + b * x + c * x ^2

I want to be able to calculate the following variables and parameters:
n, SumX, SumY, SumX^2, SumY^2, SumXY, SumX/n, SumY/n, x sigma n, y
sigma n, x sigma n-1, y sigma n-1, a, b, (c), r, r2, sey, F-
statistics, and Chi2 statistic.
Can any one provide me with such algorithms or tell me where I can
find it? Fortran or C code would also be welcome although I prefer
VBA for use in Excel.

Thanks in advance
Chuck (alias profitscalper)

>>> mislam1977@... 11/1/2005 1:28:58 AM >>> wrote

<<<
Here i have query to estimate significance p-value, as per my
knowledge when we compare two or more than two groups so we can
estimate p-value by using different statistical test according to type

of variable, is it possible that in one variable have five categories
we can estimate p-value for that for example i have grade for Quality
of life like (Grade 0, Grade I, Grade II, Grade III and Grade IV) and
responses are 10%, 17%, 17%, 23% and 39% respectively, so can i say
that Group IV is significantly higher than other, if yes so how and
which test will be use. please comment.
>>>>

It is unclear to me what you want to test.....

If you want to test whether the distribution of grades is uniform, then
you want
a one-way chi-square.

If it's something else, please let us know

Peter



Peter L. Flom, PhD
Assistant Director, Statistics and Data Analysis Core
Center for Drug Use and HIV Research
National Development and Research Institutes
71 W. 23rd St
http://cduhr.ndri.org
www.peterflom.com
New York, NY 10010
(212) 845-4485 (voice)
(917) 438-0894 (fax)

Peter Flom <flom@...> wrote:

>>> mislam1977@... 11/1/2005 1:28:58 AM >>> wrote

<<<
Here i have query to estimate significance p-value, as per my
knowledge when we compare two or more than two groups so we can
estimate p-value by using different statistical test according to type

of variable, is it possible that in one variable have five categories
we can estimate p-value for that for example i have grade for Quality
of life like (Grade 0, Grade I, Grade II, Grade III and Grade IV) and
responses are 10%, 17%, 17%, 23% and 39% respectively, so can i say
that Group IV is significantly higher than other, if yes so how and
which test will be use. please comment.
>>>>

It is unclear to me what you want to test.....

If you want to test whether the distribution of grades is uniform, then
you want
a one-way chi-square.

If it's something else, please let us know

Peter



Peter L. Flom, PhD
Assistant Director, Statistics and Data Analysis Core
Center for Drug Use and HIV Research
National Development and Research Institutes
71 W. 23rd St
http://cduhr.ndri.org
www.peterflom.com
New York, NY 10010
(212) 845-4485 (voice)
(917) 438-0894 (fax)

"Leila T. Viriña" <lengzki@...> wrote:

It depends on the objectives and limitations (e.g.
cost/budget involved) of your study.  Sampling designs
vary and you can just make manual computations.

--- sripal kumar <sripalkumar5@...> wrote:

>
> Hi guys,
>
> I was wondering if anyone can explain how to
> calculate sample size with an example (preferably
> medical statistics).  Which software can you use to
> find the sample size and what are the assumptions
> needed.
>
> thanks in advance,
>
> Sripal.
>
>
>
> Sripal Bangalore, M.D.St Lukes Roosevelt Hospital
> Center,Columbia University College of Physicians and
> Surgeons,New York, NY 10019.
>
>            
> ---------------------------------
>  Yahoo! FareChase - Search multiple travel sites in
> one click. 



     
           
__________________________________
Yahoo! Mail - PC Magazine Editors' Choice 2005
http://mail.yahoo.com

Probability questions often don't have unique answers. This is one of
them. First you have to model the situation. What assumptions are you
ready to make?

Do you assume that the date of death is independent of when we ask the
question? Do you assume that each day from here foreward is equally
likely? Does the person have a constant probability of death and each
day represents a separate trial?

We have a saying in English: There's no free lunch! That means that
you can't get an answer to a question without making some effort at
understanding different ways of interpreting the question.

It depends on the objectives and limitations (e.g.
cost/budget involved) of your study.  Sampling designs
vary and you can just make manual computations.

--- sripal kumar <sripalkumar5@...> wrote:

>
> Hi guys,
>
> I was wondering if anyone can explain how to
> calculate sample size with an example (preferably
> medical statistics).  Which software can you use to
> find the sample size and what are the assumptions
> needed.
>
> thanks in advance,
>
> Sripal.
>
>
>
> Sripal Bangalore, M.D.St Lukes Roosevelt Hospital
> Center,Columbia University College of Physicians and
> Surgeons,New York, NY 10019.
>
>
> ---------------------------------
>  Yahoo! FareChase - Search multiple travel sites in
> one click.





__________________________________
Yahoo! Mail - PC Magazine Editors' Choice 2005
http://mail.yahoo.com

Hi guys,

I was wondering if anyone can explain how to calculate sample size with an example (preferably medical statistics).  Which software can you use to find the sample size and what are the assumptions needed.

thanks in advance,

Sripal.

Sripal Bangalore, M.D.

St Lukes Roosevelt Hospital Center,

Columbia University College of Physicians and Surgeons,

New York, NY 10019.

 

---

Yahoo! FareChase - Search multiple travel sites in one click.

I think that you'll need to be more specific. Steve
Hulley's book, Designing Clinical Research, contains
simple examples. In general, power analysis isn't
something for amateurs---meaning, "Kids, don't try
this at home." Your question is analogous to a
non-physician asking, "Can you tell me how to do Botox
injections to decrease spasticity following stroke."

Advice from an experienced statistician is strongly
recommended.

SR Millis

--- sripal kumar <sripalkumar5@...> wrote:

>
> Hi guys,
>
> I was wondering if anyone can explain how to
> calculate sample size with an example (preferably
> medical statistics).  Which software can you use to
> find the sample size and what are the assumptions
> needed.
>
> thanks in advance,
>
> Sripal.
>
>
>
> Sripal Bangalore, M.D.St Lukes Roosevelt Hospital
> Center,Columbia University College of Physicians and
> Surgeons,New York, NY 10019.
>
>
> ---------------------------------
>  Yahoo! FareChase - Search multiple travel sites in
> one click.


Scott R Millis, PhD, MEd, ABPP (CN & RP)
Professor & Director of Research
Rehabilitation Institute of Michigan
Wayne State University School of Medicine
261 Mack Blvd
Detroit, MI 48201
Email: smillis@...
Tel: 313-993-8085
Fax: 313-745-9854

*********************************************************
This electronic message may contain information that is confidential and/or
legally privileged. It is intended only for the use of the individual(s) and
entity named as recipients in the message. If you are not an intended recipient
of this message, please notify the sender immediately and delete the material
from any computer. Do not deliver, distribute or copy this message, and do not
disclose its contents or take any action in reliance on the information it
contains. Thank you.

-----Original Message-----
From: Statisticians_group@... [mailto:Statisticians_group@...]On Behalf Of sripal kumar
Sent: Tuesday, November 01, 2005 4:29 PM
To: Statisticians_group@...
Subject: [Statisticians_group] Sample size

Hi guys,

I was wondering if anyone can explain how to calculate sample size with an example (preferably medical statistics).  Which software can you use to find the sample size and what are the assumptions needed.

thanks in advance,

Sripal.

Sripal Bangalore, M.D.

St Lukes Roosevelt Hospital Center,

Columbia University College of Physicians and Surgeons,

New York, NY 10019.

 

---

Yahoo! FareChase - Search multiple travel sites in one click.

At 05:21 01/11/2005 -0800, ginjac wrote:
>Dear All,
>
>Thank you for your different view points. As we moving closer, I
>think the problem seems to be more complecated.
>
>So I would like to share the occasion when I had been asked the question.
>
>One of my friends father had died exactly in his birthday. He had
>one insurance policy at that time. This question was asked by an
>insurance agent.
>
>What he wanted to know is, if I have any clear answer for finding "
>the probability that a man died in his birthday"
>
>He wants only a general, means, some man, some year,...
>
>Can we find the answer?

About 1 in 365.  It really is that simple!

Regards

Martin

The real problem is that the question is not defined.

Does it mean the

1) Probability that a particular person will die on his birthday in a
particular year (e.g. the prob. that I will die on July 2, 2006)?

2) Probability that a particular person will die on his birthday in SOME
year?

3)  Probability that some person somewhere will die on his birthday
(either in some year or this coming year)?

or what?

Peter

Peter L. Flom, PhD
Assistant Director, Statistics and Data Analysis Core
Center for Drug Use and HIV Research
National Development and Research Institutes
71 W. 23rd St
http://cduhr.ndri.org
www.peterflom.com
New York, NY 10010
(212) 845-4485 (voice)
(917) 438-0894 (fax)

>>> mislam1977@... 11/1/2005 1:28:58 AM >>> wrote

<<<
Here i have query to estimate significance p-value, as per my
knowledge when we compare two or more than two groups so we can
estimate p-value by using different statistical test according to type

of variable, is it possible that in one variable have five categories
we can estimate p-value for that for example i have grade for Quality
of life like (Grade 0, Grade I, Grade II, Grade III and Grade IV) and
responses are 10%, 17%, 17%, 23% and 39% respectively, so can i say
that Group IV is significantly higher than other, if yes so how and
which test will be use. please comment.
>>>>

It is unclear to me what you want to test.....

If you want to test whether the distribution of grades is uniform, then
you want
a one-way chi-square.

If it's something else, please let us know

Peter



Peter L. Flom, PhD
Assistant Director, Statistics and Data Analysis Core
Center for Drug Use and HIV Research
National Development and Research Institutes
71 W. 23rd St
http://cduhr.ndri.org
www.peterflom.com
New York, NY 10010
(212) 845-4485 (voice)
(917) 438-0894 (fax)

At 16:11 01/11/2005 +0900, Priyantha Wijayatunga wrote:
>Hi,
>
>Anything like 1/365 or 1/366 or 1/365.25 are wrong.
>
>If one of them is right then;
>
>for simplicity assume there are only 365 days per year.  Let's take 365
>   men (assumed to be independent individuals).
>
>Then with the simple rule of probability
>
>P[ at least one of them would die on his birthday ] = P[first person
>dies on his birthday] + P [ Second person dies on his birthday ] +
>........... + P[365th person dies on his birthday ]
>(as they are independent events)

The above is incorrect.

P(A or B) = P(A) + P(B) - P(A and B)

Regards

Martin

Peter L. Flom, PhD
Assistant Director, Statistics and Data Analysis Core
Center for Drug Use and HIV Research
National Development and Research Institutes
71 W. 23rd St
http://cduhr.ndri.org
www.peterflom.com
New York, NY 10010
(212) 845-4485 (voice)
(917) 438-0894 (fax)


>>> azizfar@... 11/01/05 12:01 AM >>> wrote
I know there are several ways of interpreting normality of the data.
Like using 95% CI and mean and SD but how can one use other measures
like Skewness and Kurtosis to our advantage.



First, the mean and sd say nothing about the normality of a
distribution.

Second, for a normal distribution, skewness and kurtosis should be 0
(depending on the program you use.....there are different formulas for
kurtosis, and you will have to check).

Third, there are a lot of different tests for normality.  None are that
great.  The problems are 1) There are too many ways to be nonnormal,
it's hard to check them all.   2) All tests that report a p-value are
sensitive to N, which you probably don't want.

Finally, the best way to check is graphical methods, in particular, the
quantile normal plot


HTH

Peter

The real problem is that the question is not defined.

Does it mean the

1) Probability that a particular person will die on his birthday in a
particular year (e.g. the prob. that I will die on July 2, 2006)?

2) Probability that a particular person will die on his birthday in SOME
year?

3)  Probability that some person somewhere will die on his birthday
(either in some year or this coming year)?

or what?

Peter

Peter L. Flom, PhD
Assistant Director, Statistics and Data Analysis Core
Center for Drug Use and HIV Research
National Development and Research Institutes
71 W. 23rd St
http://cduhr.ndri.org
www.peterflom.com
New York, NY 10010
(212) 845-4485 (voice)
(917) 438-0894 (fax)

Use this correlation matrix to answer the following:

Job performance Test 1 Test 2 Test 3 Test 4
Job Performance 1.00 .286 .350 -.410 .031
Test 1 1.00 .678 -.782 .126
Test 2 1.00 -.633 .193
Test 3 1.00 -.120
Test 4 1.00


A. Of the four tests, which is best correlated with job
performance?

B. Assume we want a test that accounts for three times as much
variance of job performance. What would the correlation between the
test (the one you selected for answer a) and job performance have to
be?


Can anyone help me with the correct solution with justification?

Thanks in advance

Use statistical packages like SAS or... I think SPSS
can also produce output like this.

--- profitscalper <aac_finlab@...> wrote:

> Hallo everybody,
>
> I am looking for computer algorithms, preferably in
> Excel VBA, for
> calculating non-linear regression parameters. I am
> particularly
> interested in the following non-linear regression
> models:
>
> - Logarithmic model: y = a + b * ln(x)
> - Exponential model: y = a * e ^ (b * x)
> - Power model:       y = a * x ^ b
> - Inverse model:     y = a + b / x
> - Quadratic model:   y = a + b * x + c * x ^2
>
> I want to be able to calculate the following
> variables and parameters:
> n, SumX, SumY, SumX^2, SumY^2, SumXY, SumX/n,
> SumY/n, x sigma n, y
> sigma n, x sigma n-1, y sigma n-1, a, b, (c), r, r2,
> sey, F-
> statistics, and Chi2 statistic.
> Can any one provide me with such algorithms or tell
> me where I can
> find it? Fortran or C code would also be welcome
> although I prefer
> VBA for use in Excel.
>
> Thanks in advance
> Chuck (alias profitscalper)
>
>
>
>




__________________________________
Yahoo! FareChase: Search multiple travel sites in one click.
http://farechase.yahoo.com

Use this correlation matrix to answer the following:

Job performance Test 1 Test 2 Test 3 Test 4
Job Performance 1.00 .286 .350 -.410 .031
Test 1 1.00 .678 -.782 .126
Test 2 1.00 -.633 .193
Test 3 1.00 -.120
Test 4 1.00


A. Of the four tests, which is best correlated with job
performance?

B. Assume we want a test that accounts for three times as much
variance of job performance. What would the correlation between the
test (the one you selected for answer a) and job performance have to
be?


Can anyone help me with the correct solution with justification?

Thanks in advance

Hello there,

I am bit busy to write. I think I rushed to the answer. Anyway, you are
partly right.

I see this as

P[ the death is on his birthday | a person dies on a day] = 1/365

assuming there are only 365 days per year. This is simple as you know if
a man dies it can be on one of 365 days.

Now we simplify this

P[(death is on his birthday) AND (man dies on a day)] /P[ man dies on a
day] = 1/365

P[man dies on birthday] / P[man dies on a day] = 1/365

Now we need to find P[man dies on a day]

Suppose average life expentency for a man is 75*365 days.

Then P[man dies on a day] = 1/(75*365)

So, P[man dies on birthday] = (1/365)*(1/75*365)

Note that again the answer is subjective we assumeed that life expetency
of a man is 75 years.

Thansk.

Priyantha



jerkos wrote:
> Hi,
> 1) where is the -P(1st person die his birthday and 2nd too) -P(1st ^ 2nd ^
> 3rd) -P(2nd ^ 3rd) etc.
> 2) question was not clear
> 3) the answer is much easer, when using 1-P(A') probability.
> If P(A) is probability that at least one of man die birthday => P(A') is
> prpb. no one die in birthday.
> P(A')=364/365 for 1 person
> P(A')=364/365*363/365 for 2 person
> P(A')=364/365*363/365*362/365 for 3 person
> etc.
>
> Tomasz
>
> ----- Original Message -----
> From: "Priyantha Wijayatunga" <spwijay@...>
> To: <Statisticians_group@...>
> Sent: Tuesday, November 01, 2005 8:11 AM
> Subject: Re: [Statisticians_group] Birthday question
>
>
>  > Hi,
>  >
>  > Anything like 1/365 or 1/366 or 1/365.25 are wrong.
>  >
>  > If one of them is right then;
>  >
>  > for simplicity assume there are only 365 days per year.  Let's take 365
>  >  men (assumed to be independent individuals).
>  >
>  > Then with the simple rule of probability
>  >
>  > P[ at least one of them would die on his birthday ] = P[first person
>  > dies on his birthday] + P [ Second person dies on his birthday ] +
>  > ........... + P[365th person dies on his birthday ]
>  > (as they are independent events)
>  >
>  > So, if you assign  P[man idies on his birthday ] = 1/365
>  >
>  > then you get
>  >
>  > P[ at least one of them would die on his birthday ] = 1
>  >
>  > So, it says certainly at least one would die on his birthday!!!!!
>  >
>  > Do you see such a thing in you life!!!!
>  >
>  >
>  > Priyantha
>  >
>  > Chirag R. Patel wrote:
>  >> 1/365.25
>  >>
>  >> ginjac <gin_jac@...> wrote:
>  >>
>  >>     Hi all,
>  >>
>  >>     I got the following question:
>  >>
>  >>     What is the probability that a man die exactly in his birthday?
>  >>
>  >>     Can we find out a solution?
>  >>
>  >>     Gin
>  >
>  >
>  >
>  >
>  >
>  >
>  >
>  >
>  >
>  > Yahoo! Groups Links
>  >
>  >
>  >
>  >
>  >
>  >
>  >
>  >
>
>
>
>
>
>
>
>
>
>
> ------------------------------------------------------------------------
> Yahoo! Groups Links
>
>     * To visit your group on the web, go to:
>       http://in.groups.yahoo.com/group/Statisticians_group/
>
>     * To unsubscribe from this group, send an email to:
>       Statisticians_group-unsubscribe@...
>      
<mailto:Statisticians_group-unsubscribe@...?subject=Unsubscribe>
>
>     * Your use of Yahoo! Groups is subject to the Yahoo! Terms of
> Service <http://in.docs.yahoo.com/info/terms/>.
>
>

Dear Colleagues,

No, this is not what is required here.  I sent an email to the group
last week indicating that you need to use Crosstabs to get the chi-
square test to compare 2 or more proportions.

The chi-square one sample test in the non-parametric tests menu is a
goodness-of-fit test, which is used in a quite different context.  I
tried it out on a file containing records for 478 elderly people who
were ever admitted to hospital as an emergency during a particular
year.  Of these, 219 were male and 259 female.  By default, this test
examines whether this sex ratio differs significantly from 1:1.  The
expected frequencies are 478 / 2 = 239 in each group, and SPSS
calculates chi-square simply as 2 * (20**2) / 239 = 3.347, df=1,
p=0.067.  This tells us that the evidence that females outnumber
males in the population from which this sample is drawn does not
quite reach statistical significance.

The null hypothesis doesn't have to be a 1:1 ratio.  Perhaps we know
that in the elderly population as a whole, there are 1.3 females to
every male.  Then the expected number of males becomes 478 * (1/2.3)
= 207.8, the expected number of females becomes 478 * (1.3/2.3) =
270.2, and we get chi-square = 1.063, p=0.303.  This would tell us
that there are slightly (non-significantly) fewer females in the
sample of 478 people, than we would expect from the known sex ratio
of the elderly population.

The test could also be applied if, say, we found the numbers with
blood groups O, A, B and AB in a series of patients diagnosed with
peptic ulceration, and we knew the proportions in the relevant
underlying population.  This example with k=4 groups would give a k-1
= 3 df chi-square goodness-of-fit test.

In each situation, you would certainly want to report the observed
proportions in the k categories, preferably with 95% confidence
intervals calculated using the first block of the spreadsheet
CIPROPORTION.XLS.  In the case k=2, suppose as above the two
categories are male and female.  Let p denote the proportion male,
with 95% confidence limits L and U.  Then you could report that the
male:female ratio in the sample is p/(1-p), with 95% confidence
limits L/(1-L) and U/(1-U).

Hope this helps.


--- In Statisticians_group@..., "Isaac Dialsingh"
<consult@t...> wrote:
> The chi-square one sample test performs this test. I beleive you
can find it
> in the non-parametric tests menu.
>
> Isaac
> ----- Original Message -----
> From: "sohail" <sohail@s...>
> To: <Statisticians_group@...>
> Sent: Thursday, October 27, 2005 2:17 AM
> Subject: [Statisticians_group] Congratulations and Need help
>
>
> > Congratulations Kundu on starting this gruoup. Dear fellows, I
want to
> > test the hypothesis of equality of two population proportions
using
> > SPSS. I am unable to find the command, someone who knows it, help
me
> > in this regard.

Hallo everybody,

I am looking for computer algorithms, preferably in Excel VBA, for
calculating non-linear regression parameters. I am particularly
interested in the following non-linear regression models:

- Logarithmic model: y = a + b * ln(x)
- Exponential model: y = a * e ^ (b * x)
- Power model:       y = a * x ^ b
- Inverse model:     y = a + b / x
- Quadratic model:   y = a + b * x + c * x ^2

I want to be able to calculate the following variables and parameters:
n, SumX, SumY, SumX^2, SumY^2, SumXY, SumX/n, SumY/n, x sigma n, y
sigma n, x sigma n-1, y sigma n-1, a, b, (c), r, r2, sey, F-
statistics, and Chi2 statistic.
Can any one provide me with such algorithms or tell me where I can
find it? Fortran or C code would also be welcome although I prefer
VBA for use in Excel.

Thanks in advance
Chuck (alias profitscalper)

Hi all,

 

I got the following question:

 

What is the probability that a man die exactly in his birthday?

 

Can we find out a solution?

 

Gin

---

Yahoo! FareChase - Search multiple travel sites in one click.

Hi again,

one correction in 3)
364/365 for 1
(364/365)^2 for 2
(364/365)^3 for 3
etc.

Tomasz
----- Original Message -----
From: "Priyantha Wijayatunga" <spwijay@...>
To: <Statisticians_group@...>
Sent: Tuesday, November 01, 2005 8:11 AM
Subject: Re: [Statisticians_group] Birthday question


> Hi,
>
> Anything like 1/365 or 1/366 or 1/365.25 are wrong.
>
> If one of them is right then;
>
> for simplicity assume there are only 365 days per year.  Let's take 365
>  men (assumed to be independent individuals).
>
> Then with the simple rule of probability
>
> P[ at least one of them would die on his birthday ] = P[first person
> dies on his birthday] + P [ Second person dies on his birthday ] +
> ........... + P[365th person dies on his birthday ]
> (as they are independent events)
>
> So, if you assign  P[man idies on his birthday ] = 1/365
>
> then you get
>
> P[ at least one of them would die on his birthday ] = 1
>
> So, it says certainly at least one would die on his birthday!!!!!
>
> Do you see such a thing in you life!!!!
>
>
> Priyantha
>
> Chirag R. Patel wrote:
>> 1/365.25
>>
>> ginjac <gin_jac@...> wrote:
>>
>>     Hi all,
>>
>>     I got the following question:
>>
>>     What is the probability that a man die exactly in his birthday?
>>
>>     Can we find out a solution?
>>
>>     Gin
>
>
>
>
>
>
>
>
>
> Yahoo! Groups Links
>
>
>
>
>
>
>
>