Math Forum / Mathematics / Statistics / February 2010

A very simple Monte Carlo exercise
PC slowness and unsophisticated random numbers generators are absolutely serious (unsurpassable) drawbacks to any professional Monte Carlo project.
HOWEVER
It’s easy to show that even using a slow program language as QBASIC one is able to provide ...

I would appreciate any clarification on this question.  Let's say I
have 52 data points each representing the percent of families living
in poverty in each State.  I have ID'ed the 80th percentile and let's
say it's 14.5% and I'd like to adjust it downward by the sampling

I've performed some test of a system (it doesn't matter what the test
does). In short the system enhances a given estimate.
The test is performed with a set of parameters chosen manually.
 - The system tests e.g. 20 different "positions".

We are having an argument at work and don't have the statistical chops
to back up my gut instincts.
Hopefully, someone could help me determine if I am totally offbase/
misremembering/or just wrong.

I need to double-compute the estimation of cumulative incidence of an
event over the periods Month M0 to month M12, M0 to M24, M0 to M36,
etc ... based on Kaplan-Meier estimate of the Survival function
(Cumulative incidence = 1 - S(t)). This is a randomized 10-year

Any input on the following would be greatly appreciated and the
specifics in the following information are made up only to illustrate
the issue and my question I have about the applicability of Principal
Components Analysis (PCA). To anyone from TX, I only picked TX due to

I have 2 questions that are related to each other:
1. I have a set of measurement scales that are highly correlated with
each other. I have tried to test the goodness-of-fit  of my model
using SEM and the Maximum Likelihood method. SEM analyses can be

Influence of r on the N(0,1)/N(m,sd)  fractiles.
Correlated normal data ratios are not infrequently found in mathematical models.
This (tinny) note has the purpose to find out the usually interesting fractiles, 0.95, 0.975 and 0.99, of the ratios between the normal standard and ...

This is my question :
Supposing that we have a number of gathered probabilities P(X=x1) ....
P(X=xn). Is there a method to estimate or define a probability
distribution function that fits the collected probabilities ?

In logistic regression, why is wald-chi square used to interpret
paremater
estimate (coffeceints of regression) as in output below? What is Wald
chi-

I have the following data and I would like to know if the frequencies
are evenly distributed across the plots. How I can prove that the
frequency of plots are significantly different from each other.
Plots  Frequency

Assume that X_1, X_2 and X_3 are three normally distributed random
variables with standard deviations s_1, s_2 and s_3. It is known that
the random variable Y=a_1*X_1+a_2*X_2+a_3*X_3 (for some fixed scalars
a_1, a_2, a_3) is also a normal random variable, whose variance is

In logistic regression, why is wald-chi square used to interpret
paremater
estimate (coffeceints of regression) as in output below? What is Wald
chi-

Have you ever seen a case of an author publishing a truly original
work, but adding co-authors that he has never worked with, just to
increase the chances of having his paper accepted in a journal, or to
make his paper more credible?

I have a set of questions/ variables representing some latent
constructs in a questionnaire that I want to explore using factor
analysis. The skewness results of some of the questions was greater
than 1/ -1 e.g. -1.35. This indicates that these variables violate the