Questions tagged [nonparametric]
Use this tag to ask about the nature of nonparametric or parametric methods, or the difference between the two. Nonparametric methods generally rely on few assumptions about the underlying distributions, whereas parametric methods make assumptions that allow data to be described by a small number of parameters.
2,126
questions
42
votes
2
answers
6k
views
Is there a reliable nonparametric confidence interval for the mean of a skewed distribution?
Very skewed distributions such as the log-normal do not result in accurate bootstrap confidence intervals. Here is an example showing that the left and right tail areas are far from the ideal 0.025 ...
5
votes
3
answers
133
views
Non-parametric one-sample mean test for a bounded variable (based on Chebyshev's inequality?)
The problem
I have $x_1, \ldots , x_n$ i.i.d. draws from r.v. $X$ such that $0 \leq X \leq 1$, but I can't make any other assumptions about the distribution.
I want to test the null hypothesis that $E(...
2
votes
0
answers
55
views
expectation functional property
I'm trying to solve the second statement of the following exercise. It is Exercise 2.5 of "All of Nonparametric Statistics, Larry Wasserman".
My try:
\begin{align}
|T(F)-T(G)| &= \left|\...
1
vote
0
answers
32
views
Combining Data from Multiple Participants for Statistical Analysis of a Facial Expression Memory Task
Context of the Query: I am working on a project involving a memory task for facial expressions, consisting of two phases:
Encoding Phase: In this phase, participants are exposed to 8 virtual ...
1
vote
0
answers
25
views
Paired Non-parametric test with more than 3 groups but just two timepoints
I may not have found the right term to search for, so I am asking what might be a straightforward question:
Assume we have an experiment with paired non-normal distributed data, called "Pre" ...
0
votes
0
answers
26
views
Local linear kernel regression
It is know that the prediction for a given point $x$ is given by:
$$\hat{f}_h(x) = \hat{\beta}_0(x)$$
where
$$\hat{\beta}(x) = \arg\min_{\beta_0, \beta_1}\sum_{i=1}^nK\left(\frac{x - x_i}{h}\right)(...
3
votes
1
answer
423
views
Theoretical justification of Parametric bootstrap?
I've been reading about bootstrap, and while it's relatively easy to find theoretical results (consistency and higher-order correctness) for the nonparametric bootstrap (e.g., Asymptotic Statistics by ...
0
votes
0
answers
22
views
Statistical Significance Testing for Nested Cross-Validation in ML Experiment
I am currently working on an ML experiment where I use a nested 5-cross validation procedure and obtain a NDCG@10 scores for each test user. I am comparing 6 different ML algorithms and have data for ...
2
votes
1
answer
408
views
Most appropriate correlation test for continuous and binary variables for non-normally distributed dataset with a high sample size
I have a dataset with N ≈12800 with two types of variables: independent continuous (distances in m) and a dependent binary variable (yes and no), associated with each distance. I have to test the ...
1
vote
2
answers
351
views
Kruskal-Wallis Grouping factors?
I am writing a report looking at human sexual selection and how there is a difference in the preferred age of potential mate between sexes. I collected 104 lonely hearts advertisements and organised ...
3
votes
0
answers
42
views
What is the minimum Pearson sample correlation given a perfect sample rank correlation with no ties? [duplicate]
Let $(X, Y)$ be a random sample of finite size $n$ from a bivariate continuous distribution with unknown parameters $(\rho_{XY},\mu_X,\mu_Y,\sigma_X,\sigma_Y)$. Assume observed values are real numbers ...
3
votes
0
answers
90
views
Best Test For Generalized/Nonparametric Behrens-Fisher Problem
First, because there seems to be a confusion of what the generalized Behrens-Fisher problem is a description adapted from [1].
We have $X_1,\ldots,X_m$ i.i.d. from distribution $P$ and $Y_1,\ldots,...
3
votes
2
answers
204
views
sample size in chi-squared test
The chi-square test of independence is a type of non-parametric test, but in cases of small sample sizes, the Fisher's exact test should be used instead. My understanding of non-parametric methods is ...
22
votes
7
answers
2k
views
Isn't it problematic to look at the data to decide to use a parametric vs. non-parametric test?
I've seen in some instances of people mentioning that using a parametric vs. non-parametric approach may be decided by looking at the data. For example this question: nonparametric vs. parametric
Isn'...
2
votes
2
answers
441
views
Survivor function for log-normal from flexsurvreg output
I am trying to plot/generate a survival curve in Excel using the output from flexsurvreg in R. The below is a snapshot from R with the corresponding estimates (y axis values) for the time (x axis ...
0
votes
0
answers
26
views
Statistical test for unequal sample sizes of repeated measures (non-parametric)
I'm looking to understand the differences between various sources (50) that evaluate an event based on 8 ordinal scales (varying from 2 to 4 values) associated with a discrete score between 0 and 10, ...
4
votes
1
answer
2k
views
Friedman's test to identify best of multiple classifiers on multiple domains
I have several classifiers $f_i\ (i=1, \cdots, N)$ and calculated performance measures on multiple domains $(D)$ for each. Thus, there are $N \times D$ values.
I want to find out (increasing ...
11
votes
3
answers
14k
views
Friedman test and post-hoc test for Python
In my dataset, I have five (ordinal) groups with an x-amount of measurement. Because homoscedasticity is violated, I performed the Friedman chi-square test to see if there are any statistical ...
0
votes
0
answers
33
views
How do I get from integral to computational formula of Cramér-von Mises statistic step by step?
My question is related to the calculus procedures behind Cramér-von Mises test.
How do I get from integral:
to computational formula:
of Cramér-von Mises statistic, step by step?
I understand the ...
1
vote
1
answer
118
views
Deriving Sample version of Anderson Darling test statistic from the theoretical version
In literature, I have seen two types of Anderson-Darling test statistic. One is expressed as
$A_T^2 = n\int_{-\infty}^{\infty}\frac{(F_n(x)-F(x))^2}{F(x)(1-F(x))}dF(x)$ and the other is given by $A_s^...
1
vote
1
answer
309
views
Why can the null and alternate hypotheses of McNemar's test be written in terms of dependance/independance of tests?
Consider a study to examine whether food frequency questionnaires and three-day food diaries are equally likely to label a women as consuming less than the RDA of calcium. Because calcium intake can ...
103
votes
6
answers
99k
views
Kendall Tau or Spearman's rho?
In which cases should one prefer the one over the other?
I found someone who claims an advantage for Kendall, for pedagogical reasons, are there other reasons?
1
vote
1
answer
62
views
What does "not relying on any parametric functional form to be postulated" mean?
I am reading the introduction section to a ml-book where I came across this sentence which I do not quite understand.
The Gaussian process is a non-parametric and probabilistic model for nonlinear ...
0
votes
0
answers
14
views
When is Tukey post hoc test result identical to Games Howell post hoc test for Welch ANOVA?
In what situations is the Tukey post hoc test identical to the Games Howell post hoc test? I have an ANOVA analysis where HOV is violated and I have a statistically significant Welch ANOVA result. I ...
0
votes
0
answers
80
views
Non-Parametric Two Way Within-Subjects ANOVA
I have a dataset with 4 individuals that are measured twice in each of the 5 groups (so in total 40 observations).
Subject
ID
Group
Value
1
1
A
45
1
2
A
62
1
1
B
70
1
2
B
37
...
...
...
...
4
2
...
4
votes
2
answers
2k
views
Student's t vs Mann-Whitney U
I want to see if there is any difference in the number of dangerous permissions requested by free vs paid apps in Android store. I have equal sample sizes of 1900. When I plot the data they are both ...
2
votes
1
answer
31
views
Does the sign test only work on location families?
That is,
if $G$ is the distribution of the sample, what does it test:
\begin{align}
\mathcal H_0 : G(x) = F(x) && \mathcal H_1 : G(x) = F(x - \theta), \theta \neq 0
\end{align}
where $F$ is ...
1
vote
1
answer
241
views
Which statistical test should I use if the assumptions of a 2-way ANOVA are not met?
My study design consists of two factors (one with 2 levels, the other with 6) and a continuous response variable. In order to analyze the influence of both factors on the explanatory variable I built ...
0
votes
1
answer
575
views
Closed form equations for simple linear regression estimators
I'm learning specifically about different forms of simple linear regression including ordinary least squares, median absolute deviation, and Theil-Sen. I have no background whatsoever in linear ...
2
votes
1
answer
1k
views
Confidence intervals for non-parametric statistics
I have a positive random variable $X$ (say, price or latency or energy) and I want to be able to say something like
I am 95% confident that $P(X<42)>0.9$ (percentile) or $P(X<22)\ge\frac12$ (...