Completely off topic BS thread!!!

[Viper_ed]

Clark, the stripper is sitting here, talking..and I wish I had a brick to chuck at her head to shut the bitch up!

[Jper]

Why wasnt I thinking of this when she was trying on outfits

[turbo83coupe]

Jper- was this the taller one? lol or the one with the kid?

[Jper]

Jper- was this the taller one? lol or the one with the kid?

LMAO the taller one.

[Troll]

anybody watch "it's always sunny in philadelphia" last night?

that show is f'd up to the max. freaking hilarious.

when they blew that guy's head clean off, one of them said, "check his pulse".

[Troll]

troll5000 lives

[2001GTTT]

keep job, sell stang and HD, and get Terminator....but it better get twin turbos. Oh, and buy my bike too!

[Troll]

blub blub blub

[Martin0660]

Who the **** cares about the Anderson-Darling test statistic or the Chi-squared test statistic, or even what the distortion risk measure of S(x) is..... I mean, there are practical uses to things such as Buhlmann credibility, and many of the other topics - but i really don't know why they find it necessary to test such obscure topics...

Wow, I never thought someone would find AD score obscure. I use it A LOT to verify fit of statistical data in MiniTab (although it calculates the AD score).

I am also required to use a CHI test for true Design For Six Sigma (DFSS)efforts, but I dont personally put a lot of credibity into it. It's more of a way to make you data LOOK like it fits than anything in my world.

I'm getting to a bit of a power user of Mintab as of late as all of our design projects are now required to have at least basic elements of Six Sigma design, and we keep phasing in more as we go along. The good thing abour minitab is tat it does the testing for you and reports the scores. If you want ot see how minitab derived the formulas it's all there, and will show the work.

So to your question of who cares, I geuss I do ;)

Bob Myers

[MBH]

troll5000 lives

say what???

[SonofaBish]

Wow, I never thought someone would find AD score obscure. I use it A LOT to verify fit of statistical data in MiniTab (although it calculates the AD score).

I am also required to use a CHI test for true Design For Six Sigma (DFSS)efforts, but I dont personally put a lot of credibity into it. It's more of a way to make you data LOOK like it fits than anything in my world.

I'm getting to a bit of a power user of Mintab as of late as all of our design projects are now required to have at least basic elements of Six Sigma design, and we keep phasing in more as we go along. The good thing abour minitab is tat it does the testing for you and reports the scores. If you want ot see how minitab derived the formulas it's all there, and will show the work.

So to your question of who cares, I geuss I do ;)

Bob Myers

haha

Well, i guess I can see the purpose of the Anderson Darling in a manufacturing basis, but i'm not in that business. I'm in the business of assessing risk and putting a price tag on it... For me, the buhlmann methods develop a credibility factor for the data being estimated and that's a pretty self-explanatory term. Estimation/prediction of the next, unknown event can then be much more precise...

honestly though, these exams don't exactly always test stuff we will use practically - so my biggest gripe is that its a reasonably difficult test to do by hand on these exams. They don't exactly give you easy distributions to run the test on - its a much more powerful test than the Chi-square or the Kolmogorov-Smirnov test, but when using the Empirical distribution functions that the SOA likes to use, such as the Pareto, Weibull, and Normal, AD takes too long to compute - or atleast longer than what I consider a reasonable amount of time for a 40 question, 4 hour exam.

[Martin0660]

its a much more powerful test than the Chi-square or the Kolmogorov-Smirnov test, but when using the Empirical distribution functions that the SOA likes to use, such as the Pareto, Weibull, and Normal, AD takes too long to compute - or atleast longer than what I consider a reasonable amount of time for a 40 question, 4 hour exam.

So I ASSume that you have to manually calculate the analysis as part of these exams?? Have you ever tired / worked with mintab or northwest? With mintab I can pull a data set from Excel, do pareto in 3 minutes, and do a distribution 6 pack that will look at normal, weibul, etc and give me a AD score in 10 minutes :D What you really do with the data from there is always the trouble, but it will get you pointed in the right direction.

Effectively, what we are doing with 6 sigma tools are the same as what you describe. We are looking to predict failures at a rate about 3 parts per billion, and with DFSS, on a product that doesent even exist. If applied correctly, it works well.

Bob Myers

[Troll]

say what???

I was drinking and thinking out loud (more than normal). It was way TMI for y'all to handle... like insider trading or something. Luckily, it looks like Clark is the only one who saw it. Fortunately, his mind resets like a trip-meter every time his pecker points north checking out the latest and greatest hottie.

[Martin0660]

Luckily, it looks like Clark is the only one who saw it.

Nah, I saw the list too. I like the last three options best :D

Fortunately, his mind resets like a trip-meter every time his pecker points north checking out the latest and greatest hottie.

Fortunately for you, reading that rebooted my mind and gave me a case of the heebee jeebee's that wiped out anyhing left to do with the subject.

Bob Myers

[How Much?]

haha

Well, i guess I can see the purpose of the Anderson Darling in a manufacturing basis, but i'm not in that business. I'm in the business of assessing risk and putting a price tag on it... For me, the buhlmann methods develop a credibility factor for the data being estimated and that's a pretty self-explanatory term. Estimation/prediction of the next, unknown event can then be much more precise...

honestly though, these exams don't exactly always test stuff we will use practically - so my biggest gripe is that its a reasonably difficult test to do by hand on these exams. They don't exactly give you easy distributions to run the test on - its a much more powerful test than the Chi-square or the Kolmogorov-Smirnov test, but when using the Empirical distribution functions that the SOA likes to use, such as the Pareto, Weibull, and Normal, AD takes too long to compute - or atleast longer than what I consider a reasonable amount of time for a 40 question, 4 hour exam.

Ok guys,, just curious and dont want to sound ignorant, but i am! WTF do you guys use this stuff for? Like what the heck do you do for a living? Sounds like some crazy mind blowing stuff that my mind wouldnt even begin to understand

[SonofaBish]

So I ASSume that you have to manually calculate the analysis as part of these exams?? Have you ever tired / worked with mintab or northwest? With mintab I can pull a data set from Excel, do pareto in 3 minutes, and do a distribution 6 pack that will look at normal, weibul, etc and give me a AD score in 10 minutes :D What you really do with the data from there is always the trouble, but it will get you pointed in the right direction.

Effectively, what we are doing with 6 sigma tools are the same as what you describe. We are looking to predict failures at a rate about 3 parts per billion, and with DFSS, on a product that doesent even exist. If applied correctly, it works well.

Bob Myers

Haha, yes that's correct - I have to understand and be able to calculate everything. I have a pencil, a test booklet, and a BA-II plus calculator ... They don't even allow us to have a graphing calculator for goodness sake... But that's why they're considered the toughest professional exams out there... and If you don't understand the nitty-gritty behind the methods, and just let a computer do it all the time, then (for me) that's worthless... But obviously, on a daily basis, i'm sitting behind a computer also :) These exams are just to get my credentials

Ok guys,, just curious and dont want to sound ignorant, but i am! WTF do you guys use this stuff for? Like what the heck do you do for a living? Sounds like some crazy mind blowing stuff that my mind wouldnt even begin to understand

It sounds to me that Mr. Myers does something along the lines of quality control engineer ..

Myself, I am an actuary. My job is to do studies, and apply statistical methods in order to predict the financial impact of future, contingent events. Basically that is, analyze and evaluate the probability of an event happening, and how much that possible event will cost a company, society, or individual...

I work for Highmark Blue Cross Blue Shield, and do all of the pricing for our health insurance plans. Previously, i spent a year with Mercer H.R. Resources, where we did pension valuation - determining the amount of money that was required by the gov't to be put back now, in order to pay future pension obligations.

[2001GTTT]

I was drinking and thinking out loud (more than normal). It was way TMI for y'all to handle... like insider trading or something. Luckily, it looks like Clark is the only one who saw it. Fortunately, his mind resets like a trip-meter every time his pecker points north checking out the latest and greatest hottie.

hah...i wish.

[somethingclever]

In college I weighed 165 lbs and once ate 61 wings at wing nite @ Quaker Steak and Lube. I tried to keep up with this 300lb dude that was there.....he ended up with over 70...bastard.


I almost threw up in the car on the way home....it was VERY close.


How many wings have you ate in one sitting? (trimmed down to the bone...none of this two bites and your done stuff)

[03 OW SVT]

In college I weighed 165 lbs and once ate 61 wings at wing nite @ Quaker Steak and Lube. I tried to keep up with this 300lb dude that was there.....he ended up with over 70...bastard.


I almost threw up in the car on the way home....it was VERY close.


How many wings have you ate in one sitting? (trimmed down to the bone...none of this two bites and your done stuff)

48 of dem little SOB's in 15mins. when i was young and dumb.

[94svtcobra]

scs wing eating contest? haha