You can work as a Lean Sigma coach here
Anonymous replied to the topic Multiple Regression Best Practise – Standardization in the forum General 9 years, 9 months ago
Thank you very much for your reply.
My comment on normally distributed predictors was not related to regression assumptions but rather, if I standardize by subtracting the mean and dividing by stdev (along with a standardized response), I could then say that a coeff = 1 means that a one stdev change in predictor would be related…[Read more]
First a correction – the center line is not the median, it is the mean.
Second, the control limits are calculated using the Shewhart formulas to estimate short term (within) standard dev (MR-bar/d2), which is what you want to use (not overall stdev).
So yes you can divide the chart into +/- 2, +/- 1 zones and use WECO and Nelson.
But you would…[Read more]
Anybody notice that in all the postings in this tread that Stan has just made smart comments the entire time without any meaningful contribution?Way to go buddy. If only they could begin cloning you our entire economy would be saved!!!!Save yourself the quick 1 line reply Stan… I dont plan on returning.
Here is a link to view the 30 Rock program:
Looks like MVT didn’t save the day for Circuit City. Sorry Charles, it looks like MVT now includes failed companies, just like SS. There goes your anti SS argument.
Monday, November 3, 2008 As of 9:13 AM EST
The Wall Street Journal
Circuit City to Retrench
By KEVIN KINGSBURY
Ailing consumer-electronics chain Circuit City Stores Inc. s…[Read more]
I did some quick and dirty stats on it. There’s been an average of 41 posts per day in the past 4 weeks.
Then I searched the archives and found post numbers from two threads that I was previously involved in — one about NASA sigma scores in March 04 and one about Enron in April 06. The difference between those post numbers and the dates, in…[Read more]
Well stated Chad.
I suspect that LG doesn’t want to drill in Anwar because he thinks it will result in the state of Florida being washed away in about 6 years or that it might also result in glaciers in Philadelphia.
This is consistent with the fact that people of LG’s ilk almost always seem to be the ones who don’t think that a plane hit the…[Read more]
Liberal Greenie wrote, ” Listen to others and engage in a dialog. Thats what has been missing in the US for the last few decades. We talk at each other not to each other.”
Well LG, I agree with the sentiment but the correct place to post it would be on the Huffington Post, Moveon.org or the Daily Kos… The venom that comes from…[Read more]
What are you trying to determine? Confidence Interval for Mean? Power for One-Sample t? Two-Sample t? ANOVA? Factorial DOE? Equal Variance? Are you comparing the exponential scale parameter (lambda)?
Are you able to transform the data to normaility using Box-Cox or Johnson?
Welch’s ANOVA is more powerful than Mood’s Median or Kruskal-Wallis. It is very similar to the two-sample t-test for unequal variances. We don’t tell people to jump to Mann-Whitney in the two sample case, so why would we do that in the multiple group case?
The answer may be that it’s not in the software that you are using, but if you look…[Read more]
Are you with a PD? Did you get your training from MU?
You are correct that the technically correct tool is Poisson Regression, but Linear Regression with reasonable residuals especially with larger counts should be ok.
The difference between your regression model and correlation are likely due to collinearity in the Xs. Check the VIF…[Read more]
- Load More