Questions tagged [anderson-darling]

The A-D test assess the fit of an empirical (observed) distribution to a theoretical one. It is an especially powerful option when the theoretical distribution is the normal. The A-D test can also be used to see if several empirical distributions come from the same population.

Filter by
Sorted by
Tagged with
1
vote
0answers
29 views

Anderson-Darling test: p-value reliability when testing fitted distribution

I have a question regarding the A-D test, and perhaps goodness-of-fit tests altogether. I fitted a dataset to a long list of distributions. According to A-D, a Wakeby distribution provides the closest ...
1
vote
0answers
19 views

Can you use a test statistic like Anderson–Darling for parameter estimation?

I was looking at the Wikipedia article for maximum spacing estimation and this got me thinking. The idea behind this method is that if you know the true distribution of the data, then its CDF should ...
0
votes
0answers
17 views

Unexpected behaviour of Anderson Darling test for normality for low sample size

I’m testing the power of the Anderson Darling test to detect deviations from standard normal distributions. I’m calibrating the AD test statistic numerically using python and I noticed an unexpected ...
1
vote
1answer
30 views

Two-sample Kolmogorov-Smirnov, Anderson-Darling, and after?

I am trying to investigate the differences between two datasets. Each one is composed of 20 000 elements obtained from a Monte-Carlo method whose implementation slightly differs in both cases. I'd ...
0
votes
0answers
12 views

Sample Size Guidelines for KS-Test and AD-Test

so I've been looking into the Anderson Darling and Kologorov-Smirnov test because I have datasets with a large number of observations (~6000) and would like to fit distributions to them. They are ...
0
votes
0answers
16 views

Issue with two-samples Anderson-Darling test with same dataset (scipy anderson_ksamp)

I am computing several two samples Anderson-Darling statistics using the Python scipy.stats.anderson_ksamp function (both samples have $N = 20000$ values), and I ...
1
vote
0answers
14 views

P value calculation for Anderson Darling normality

For p-value calculation I use (in c++) following equations I found on the internet ...
0
votes
0answers
10 views

Power of Anderson-Darling with known vs unknown parameters

I recently ran some AD tests for normality on some data. I ran AD in two modes: (i) With specified (mean, sd) of the normal (ii) With unspecified (mean,sd) These correspond to cases 0 and 3, ...
1
vote
0answers
20 views

Highly fluctuating value of test statistic in Anderson Darling test

I am trying to verify whether my sampled data is from a population with a Gamma density function. My plan is to do this by means of the Anderson-Darling test. Before doing so, I thought it would be ...
0
votes
1answer
40 views

Hypothesis test for arbitrary distribution

Given a sample of $N$ observed values I'd like to test the null hypothesis that they arose from an arbitrary PDF (for which I have the analytical form). There are tests in place that can handle some ...
5
votes
1answer
271 views

Normality test for discrete values of a continuous variable

I have a dataset with several hundred pH measurements from a factory line. This dataset will be used to infer process capability using Minitab. On a physical basis, pH is a continuous scale (being ...
1
vote
0answers
312 views

testing for continuous uniform distribution in python

I'm setting up a statistical experiment. I am investigating whether a list of real values follows a continuous uniform distribution. In scipy, there is kstest to do it, I find it not robust after ...
0
votes
1answer
44 views

Checking the normality of a large sample?

If I want to analyze a large sample size (N = 50.000) of continuous data ($ revenue) from an A/B test, what would then be the best way to check for normality? Thanks in advance!
2
votes
0answers
49 views

Why do I get different results of goodness of fit test using AD tests from different R libraries?

I am trying to find the best method for goodness of fit while testing my data with different models of distributions. I need to use AD test. So I have compared results using two different ...
4
votes
1answer
72 views

Confusion about Anderson-Darling test

The Anderson-Darling test statistic is $$ A^2 := n \int_{-\infty}^{\infty} \frac{(F_n(x)-F(x))^2}{F(x)(1-F(x))} \, \mathrm{d}F(x) $$ We know that the samples $$ Z(x) := \frac{\sqrt{n}(F_{n}(x) - F(x))}...
0
votes
0answers
68 views

Comparing two-sample statistical test results: Kolmogorov-Smirnov and Anderson-Darling

I am not sure how to interpret a given Anderson-Darling criterion as computed from SciPy. Suppose, for example, I have a Kolmogorov-Smirnov (KS) statistic of 0.60 and an Anderson-Darling (AD) ...
5
votes
1answer
496 views

Kolmogorov Smirnov test vs. Anderson Darling test

I learned that Kolmogorov Smirnov loses sensitivity (power) in the tails, thus it is not adequate for testing goodness-of-fit of fat tailed distributions. However, Anderson Darling test is more ...
4
votes
1answer
744 views

Why do two implementations of the Anderson-Darling test produce such different p-values?

I want to check the normality assumption of my series I am trying to figure out the way is computed the p-value of the Anderson Darling in R packages, goftest and <...
4
votes
1answer
787 views

Alternatives to 2-sample Kolmogorov-Smirnov Test?

I have 2 samples with numerous repeating numbers like this: Sample 1: 1,1,1,1,1,1,2,2,2,2,2,2,3,3,3,3,3,4,4,4,4,... Sample 2: 1,1,1,2,2,2,3,3,3,3,3,3,3,3,3,3,4,4,4,4,4,... and I would like to ...
0
votes
0answers
95 views

Using Anderson Darling test statistic vs p value to infer Goodness of Fit

I have been reading articles recently that discourage the use of p value for hypothesis testing (https://www.sciencenews.org/blog/context/experts-issue-warning-problems-p-values) Rather than using a ...
1
vote
0answers
188 views

How to test for conditional independence with mix of categorical and continuous data

I wish to test for conditional independence between $X$ and $Y$ given $Z$. However $X$ is continuous, while $Y$ and $Z$ are categorical variables. My idea is to test if $P(X|Z = z) = P(X|Y = y, Z = z)...
2
votes
0answers
217 views

Anderson-Darling test for any distribution with unknown parameters

I have been spending the last two days trying to construct a solid answer to the question: What is the distribution that best describes my sample? I did the following: Plot the histogram (nomalized ...
0
votes
1answer
345 views

Is the Anderson-Darling test similar to the Kruskal-Wallis test to compare more than 2 groups?

I would like to compare more than 2 groups (their dependent variable is not distributed normally). Can I use the Anderson-Darling test like I would use the Kruskal-Wallis test? I have no specific ...
0
votes
0answers
86 views

Is there an equivalent to the Anderson-Darling test for categorical data?

I have categorical data of the following form: |Total|Pos|Neg In | 6 | 4 | 2 Out | 11 | 4 | 5 There are two categories: In/Out and Positive/Negative. All ...
1
vote
0answers
63 views

Kolmogorov-Smirnov test with PDFs instead of data points?

I don't have a background in statistics, so I don't know exactly how to phrase my question, and I am likely misusing some of the statistic lingo. I have a model which I am fitting to 30 sets of data,...
1
vote
1answer
72 views

Checking closeness of a mixture weibull and a mixture normal distribution to sample data

I have a sample data whose histogram shows a two peak distribution. So I generated two distributions which is the Mixture (average) of two (1) two weibull distributions (2) two normal distributions I ...
2
votes
1answer
243 views

Interpretation of the test statistics of ks.test, ad.test, and wilcox.test

I am currently writing my Master's Thesis and I have a question in regards to the Kolmogorov-Smirnov test (ks.test in R), Anderson-Darling test (ad.test in R), and Wilcox/Mann-Whitney U test (wilcox....
2
votes
1answer
1k views

Can we compute P-value of Anderson Darling Test with AD statistic without using the tables given

I am fitting my experimental data with different distributions, I am computing Anderson Darling statistic for my data and theoretical distributions. I want to compute P value from Anderson Darling ...
0
votes
1answer
92 views

How to find the samples that contribute to the differences between two distributions?

I have two very large distributions, I did anderson-darling test on these two distributions and the result shows they're from different populations, significantly. Now I want to know which samples in ...
3
votes
1answer
1k views

Distribution specificity of the Anderson-Darling test

When fitting a (gamma) distribution to a data sample and in-turn estimating its parameters (shape/scale), a Kolmogorov–Smirnov goodness-of-fit test is no longer applicable as the distribution being ...
4
votes
1answer
1k views

R and SAS produce the same test-statistics but different p values for normality tests

I use Kolmogorov-Smirnov, Cramer-von Mises and Anderson-Darling tests to examine the normality of residuals of an OLS regression. It puzzles me that R and SAS, despite giving the same test-statistics ...
1
vote
0answers
75 views

Maximizing the p-value associated with a goodness-of-fit test

Problem (concrete version): Let $p = p(T, \mathcal{D}, \mathbf{X})$ be the $p$-value associated with goodness-of-fit test $T$ (i.e. Kolmogorov-Smirnov, Anderson-Darling, Cramér–von Mises) on ...
6
votes
0answers
1k views
19
votes
2answers
14k views

2 Sample Kolmogorov-Smirnov vs. Anderson-Darling vs Cramer-von-Mises

I was wondering what are the criteria to use Kolmogorov-Smirnov, Cramer-von-Mises, and Anderson-Darling when comparing 2 ECDFS. I know the mathematics of how each differ, but if I have some ECDF data, ...
1
vote
1answer
129 views

Calculating the confidence interval

I have a simple random sample of size 40 which talks about no. of newspapers being sold by a particular salesboy each day. Population standard deviation is not know. I did a normality test on this ...
3
votes
2answers
2k views

Anderson Darling Test on student-T distribution

I would like to do an Anderson Darling test in R using ad.test() from the package “goftest”. My null hypothesis is H0: Data follows a student-t distribution with 2 degrees of freedom. I receive the ...
5
votes
1answer
1k views

Anderson-Darling test for normality with estimated parameters

I have a set of measurements coming from a manufacturing processes. I want to test if the measurements come from a normal distribution. If I understadn correctly, it's wrong to use K-S test or A-D ...
4
votes
3answers
1k views

Is it valid to rank goodness-of-fit based on the value of the Kolmogorov-Smirnov or Anderson-Darling test statistics?

When using Kolmogorov-Smirnov or Anderson-Darling goodness-of-fit tests, is it valid to claim that, because distribution X has a lower test statistic than distribution Y, distribution X is a better ...
1
vote
1answer
833 views

Scipy.stats.anderson_ksamp negative return values for test statistic

Ok, so I've been trying to run this test on the the iris dataset to see if it flags the clusters within the data as samples that aren't from the same population. from sklearn import datasets iris = ...
5
votes
3answers
2k views

sample size of Anderson-Darling test

Why must the sample size of the Anderson-Darling test for normality be greater than 7? The function ad.test in some packages in R enforce a sample size $n>7$. ...
0
votes
0answers
24 views

Gaussian based clustering/partitioning, does it make sense with not much data?

I have a dataset about hourly aggregated mobile phone usage (#calls, #sms, #internetConnections) in one mobile cell. For example I have this data about activity at 8:00am: ...
2
votes
0answers
321 views

Use of normality test to distinguish gamma from log-normal distribution

I have random population sample data that I would like to describe using a distribution. If I plot the estimated kernel density, the data appear positively skewed and using functions in R such as ...
4
votes
1answer
3k views

Simulation of KS-test with estimated parameters

We know that the original KS-test has a limitation, requiring the Null hypothesis testing underlying distribution to be fully specified, rather than estimated. But in practice, we usually need to test ...
2
votes
0answers
186 views

Multiple Sample Equality-of-Distributions Test For Panel Data

Suppose I have repeated time observations for 800 meter sprint for a group of athletes over the course of a season. I would like to test the hypothesis that all the times are drawn from the same ...
2
votes
0answers
693 views

Anderson-Darling test in 2D time series

Suppose two time series (of light flux). The goal is to determine whether the series are from the same distribution. It is usual to use the Kolmogorov-Smirnov (KS) test in this situation. However, ...
2
votes
1answer
251 views

Testing for membership in a distribution with one single value

I came across an interesting little quirk when I was coding up a hypothesis test in Python. I was wondering why very time I tested some values for normalcy using the Anderson–Darling test, I was ...
1
vote
0answers
1k views

Anderson-Darling vs Kolmogorov-Smirnov (Many points in sample)

Previous question In my previous question (Kolmogorov-Smirnov test - reliability) I misused two-sample KS test for normality testing of one-sample data. I was given an advice that I should use one-...
4
votes
1answer
820 views

Anderson-Darling test - what is adjustment good for

I am trying to understand and implement Anderson-Darling test for normal distribution in C++. I've found Octave/Matlab implementation of Anderson-Darling test here, but the critical values differ from ...
0
votes
1answer
189 views

Finding the probability of normality via Anderson-Darling, Shapiro-Wilk , and Kolomogrov-Smirov

I have quite a few distributions that were generated by some system, and I am trying to find the probability of accordance to normality for these distributions. So, the probability of accordance to ...
4
votes
1answer
873 views

Anderson-Darling code test

I'm really rusty at statistics and I'm trying to write some C# code where I feed in a list of numbers and it tells me whether or not the numbers are normally distributed. I generated 50 numbers from ...