6
votes
Closed formula for D4 constant calculation? (Moving range chart constant)
The definition of $D_3$ and $D_4$, from "ASTM MNL7, Manual on Presentation of Data and Control Chart Analysis" is:
$D_3=1-3\frac{d_3}{d_2}$
The equation in the standard should be, $D_3=\max\left(0,1-...
- 1,622
5
votes
Within-group and between-group variation in real life processes?
These models are referred to as linear mixed effect or random effects models in general.
A search for articles in this area yield a couple of potential references:
Monitoring of Linear Profiles Using ...
- 5,573
4
votes
Quality control - non-normal distribution
It looks like you are confounding Capability and Statistical Process Control.
Shewhart Charts or Control Charts typically plot what is assumed to be Gaussian data with control limits of $\pm3s$. ...
- 1,622
4
votes
"Random variable" vs. "random value" (when translating from Russian into English)
The correct phrase in order to make statistical sense, is "random variable". And this is because it refers to the "mean line" - and random values do not have a mean, only random variables do.
If ...
4
votes
Accepted
Repeats vs Replications
I don't find the "replicate" versus "repeat" terminology to be very helpful, and it seems to conflict with common use of the word "replicate" in phrases like "...
- 98.1k
3
votes
Closed formula for D4 constant calculation? (Moving range chart constant)
I know this is a closed answer, but just to extend a little bit of the discussion, and because it's also one of my favorite subjects I would like to add an article I found.
The article it's called: ...
3
votes
Accepted
"Random variable" vs. "random value" (when translating from Russian into English)
I side with your editor. You plot the obtained values on the plot, then draw a mean (average) line, then calculate three standard deviations above and below the line to show the boundaries. There's ...
- 61.9k
3
votes
Accepted
How to find the time to failure of a component using mean failure time
What time it will fail will be a random variable, so you can't say what time it will fail.
You may be able to say something about the distribution of the time to failure, including its expectation, ...
- 287k
3
votes
One-class KNN for Quality Control
At the very heart of LOF you will find "k-distance", the distance to the k-nearest neighbor.
The idea of using the k-distance is older than LOF. And the range 10..50 may be a good choice for LOF, but ...
- 42.8k
3
votes
Accepted
Non-normal distributions vs data transformation
If you want to use parametric statistics, you assume some underlying distribution. You should therefore always try to use a model with a distribution that makes sense given the process that you are ...
- 14.2k
3
votes
What do the letter and the subscript in the Shewhart's control chart constants ($A_2, A_3, c_4, c_5, \ldots$) mean?
Note: This is a partial answer.
Wikipedia mentions Duncan, A. J., Quality Control and Industrial Statistics 4th Ed., Irwin (1974) ISBN 0-256-01558-9, p.139.
An older edition of this book contains a ...
- 566
3
votes
Accepted
Average Sample Size For Curtailed Double Sampling Plan
This comes down to an analysis of "curtailment" (which is a simplified version of sequential sampling). Let's focus on that first. Because I think your reference is incorrect, I will be ...
- 330k
3
votes
Experiment design useful when objects change / will be replaced in each trial run?
The fact that the object changes from one run to the other is not an issue. As usual, the object-to-object variability will influence your inference. If the objects come by batches, then you can use ...
- 22.3k
2
votes
How can root mean square error be used to predict logistic regression model accuracy?
If I understand correctly, the situation is that you have an estimated logistic regression model that you use in production. So you need a system of quality control that can tell you if the model ...
- 81.7k
2
votes
When is a single data point enough to be considered a trend in the data?
My wife worked as a Medical Laboratory Technologist for several years. There are many tests that are not required once the test meets a desired threshold unless other indicators exist. These values ...
- 1,622
2
votes
Quasi-experimetal Research is it necessary a control group?
There are two distinct issues here. The first is the meaning of the term "quasi-experiment". Quasi-experiments, at least in the way the term was used by Don Campbell who created it, includes numerous ...
- 1,818
2
votes
Dealing with zeros in a poisson regression
Either (as mentioned in a comment by @Glen_b) use Bayesian methods, or some kind of borrowing strength, that is, analyzing multiple data sets with a common model, with some common parameters (that can ...
- 81.7k
2
votes
Accepted
How to test for deviation from a uniform distribution?
What you propose looks OK. Since there is no estimated parameters, the number of degrees of freedom should be $10000 - 1$. Then you can find p-values from the $\chi^2_{9999}$ distribution.
But your ...
- 81.7k
2
votes
Is there a Status Machine Chart?
It's three time series, each plotting the status of a single machine over time. Instead of the value of each of the three series being indicated by a y-coordinate, it is indicated by one of three ...
- 36.1k
2
votes
If the population size was large enough that the sample size required to represent it is too large to be practical, what should be done?
One of the lovely things about sampling theory is that it demonstrates that you can make good inferences from sampling, even with a very large population and a correspondingly small proportion of ...
- 130k
2
votes
Accepted
Quantifying agreement between datasets from different methods of measurement
The Bland Altman plot is a much more nuanced tool for examining agreement between different methods. It allows you to examine bias as well as agreement. You can understand if the agreement is uniform ...
- 3,223
2
votes
How to estimate probable response times, from previous samples?
You do not give much information about these waiting times,
so this answer has to be regarded as highly speculative.
Often waiting times are well-modeled as exponential. Your
small sample $0.2, 1.0, 0....
- 57.1k
2
votes
Accepted
three sigma vs six sigma
Exactly, it's related to the number of sigmas to reach the LSL/USL from the mean of the distribution B compared to distribution A. There is an excerpt from the Wikipedia page that explains this fairly ...
- 1,043
2
votes
Accepted
Using standard deviation to calculate Control Limits for Individual Control Chart
The NIST Chapter on "Process or Product Monitoring and Control" is a useful reference on this topic. As explained there, it often is preferable to use estimates of $\sigma$ based on the ...
- 98.1k
2
votes
Accepted
How many values should one validate to achieve a desired confidence?
There are a few ways to think about this.
If you want to use the confidence interval approach, then you should also look at the "finite population correction" (Googling for that phrase ...
- 52.4k
2
votes
Difference between anomaly detection and data quality check
This is such a difficult question to answer because 'data quality checks' is so broad, and even the definition of anomaly detection is not completely clear. Moreover, to give the best suggestions we ...
2
votes
Accepted
Correct setup for a Bayesian gamma model
Yes, you are correct. To confirm, I downloaded the 2021—2022 data for Table B6: Rates of patient mortality after listing. (The healthcare application is organ transplantation.) The expected & ...
- 10.5k
1
vote
Statistical Quality Control Data Gathering
As far as using your data for standard quality control methods which require that you have Gaussian data, the results of your tests confirm that you are currently unable to reject the null hypothesis ...
- 1,622
1
vote
Statistical Quality Control p-chart
As long as the sample size is consistent, you should be using an $np$ chart instead of a $p$ chart.
The rest of your approach would be appropriate for solving this type of problem.
There is the $ ...
- 1,622
1
vote
Accepted
Control Chart - Descriptive or Inferential?
"The control chart method for variables is a means of visualizing the variations that occur in the central tendency and dispersion of a set of observations."
D.H. Besterfield, Quality Control, ...
- 1,622
Only top scored, non community-wiki answers of a minimum length are eligible