Iain Hastings

1, 3, 9 is based on experience. According to Lou Cohen in his book “Quality Function Deployment – How to make QFD work for you” early practitioners in the US used 1, 3, 5 but needed a stronger contrast between “strong” and the other impacts. “9” was adopted for that purpose on the basis that it had been seen to be used in Japan in the early 80’s.…[Read more]

Jayesh,
As you are at a fairly advanced level in your education and you have a copy of Mintab, it should be reasonably straightforward for you to search and use the statguide feature to answer your questions.
No doubt if you are lucky someone will do the cut and paste for you and add it to this thread…

The “35 of the time” should read “3% of the time” in the prior post.

Mike,Your questions at the end of your post would concern me a bit. There is no time when there is “too much data when testing for normality”. The point is that your particular example of tossing two dice does not really represent a normal distribution. Re-read BC’s response – what you have is a histogram with discrete bins (2,3,4…12). You do…[Read more]

UKSS,
I would suggest that you first construct a histogram of the data and ask what the distribution shows – for example does it look to be a single distribution or is it bi-modal or multi-modal? If there are multiple distributions working together then it is probably a reasonable starting point to try to identify and break out the individual e…[Read more]

Anoop,
Ron’s answer is not a great answer and it does not sum things up accurately or appropriately. Similarly your statement is wrong and confused.
I strongly recommend that both Ron and you take the time to read Robert Butler’s comments on this thread. He has spent considerable effort trying to remove confusion, unfortunately it appears to…[Read more]

Louis,
To construct control limits you need an estimate of the standard deviation. When the sample size is relatively small an unbiased estimate can be derived from the range of the data. A2R is simply the unbiased estimate of 3*sigma

Andrew,
The 1.96 is derived from the standard normal distribution. If you look at a set of tables Z=1.96 corresponds to 0.975 (ie 95% two sided).
You should be able to answer your questions (as stated) from this.
 

This is a conditional probability problem. You know that the probability of testing +ve intersected with the probability that you don’t have the disease is 0.05. By definition this equals the probability that you test +ve times the probability (don’t have disease/Test +ve). You should be able to work it out from there.
Another way to look at it…[Read more]

It may be that you have a quantization error which is throwing out a low p-value. This could well be the case if the data is within a narrow range (ie all at 6.xxxx).
Do as Adrian suggests. Plot on normal probability paper and do the fat pencil test. Also look to see if there are little vertical “stacks” of dots staddling or close to the n…[Read more]

#6 – The gambler was the Chevalier de Mere (Antoine Gombaud) and the rather prudish mathematician was Pascal.
The question related to finishing a game early – how to divide the winnings based on the current status and the chances of each player winning from that point.
At least I think that is the answer. Robert?

#3 – Benjamin Disraeli and MarkTwain
#7 – George Bernard Shaw said it but I had no idea that it was in response to a barb. You have me intrigued here Robert – what was the barb?
 

It took me longer to write this post than it did to find the link. More distributions than you can shake a stick at.
http://en.wikipedia.org/wiki/Probability_distribution
 

Paul,
You’ve posed several questions here which I will try to answer in turn, however, it is not entirely clear to me what you are using DPMO for.
First question: Is there a standard for DPMO calculations on a manufacturing line? The short answer is “No”. When DPMO is used, companies usually develop their own guidelines and then rigidly adhere…[Read more]

Alan,
What does the table represent. You make the statement that there is no response – in which case what are the numbers? For example what does the number “15” represent at the intersection of “fish”: and “fat”?
 

Now you’ve got me confused – I need to think about that.
But, the CLT is based on taking averages of samples from a distribution. The resulting mean of these samples forms a normal distribution with the same mean as the parent population. The variance of the sample is equal to the variance of the distribution divided by the sample size.
In the…[Read more]

Yes, under those conditions it would be. The important point for the original poster is that variances can be added but standard deviations cannot. (Hence RSS).

In a word: No.
To illustrate this consider two populations each with n=500, both with a standard deviation of 1.0. One population has a mean of zero, the other has a mean of 10.
From the definition, pooled standard deviation is equal to 1, however the standard deviation of the combined populations will be about 5. (Try this at home by generating…[Read more]

I think that you should find out what your customer needs.

Minitab actually allows for 3 different normality tests. As well as Anderson-Darling, you can select Ryan-Joiner which is similar to Shapiro-Wilk or you can choose Kolmogorov-Smirnov
To pre-empt your next question here is the description of each from the Minitab statguide.
MINITAB provides three normality tests that you can choose…[Read more]