Procedures that rely on relatively few assumptions about underlying probability distributions.

learn more… | top users | synonyms

1
vote
0answers
15 views

non normal data - bootstrap pearson r or kendall tau?

This may be a follow up on these questions: Pearson's or Spearman's correlation with non-normal data How robust is Pearson's correlation coefficient to violations of normality? Since I ...
0
votes
1answer
12 views

Test to determine correlation between organism size and concentration?

I have a gamma distribution. I am looking to determine if higher concentrations of a toxin are correlated to larger organism size. I think a non-parametric test would be best, but I'm not sure which ...
3
votes
0answers
14 views

Kolmogorov-Smirnov test: horizontal variant?

The Kolmogorov Smirnov test is based on the maximum vertical distance between the ECDFs of two provided samples. Is there a variant that checks the maximum horizontal distance?
0
votes
0answers
16 views

Nonparametric alternatives to 1-way anova?

Are there non-parametric alternatives to one-way analysis of variance? Ideally, I am looking for something that is easy to understand and can be done in SPSS.
0
votes
0answers
22 views

Non parametric clustering by discretizing continuous IV

My final target is to develop a predictive model for a rate (fraction) DV. The DV showed bimodality and I have no variable that separates the two modes. Hence I created an IV using two observed IVs ...
1
vote
0answers
13 views

Can I use a Spearman Rank-Order Correlation test to check the impact of potential confounders in my data?

I have a relatively basic statistics knowledge, so I apologise in advance if this question doesn't fully make sense or if I'm totally off-base with how I'm trying to analyse it. I've got a project ...
2
votes
1answer
37 views

What is “Targeted Maximum Likelihood Expectation”?

I'm trying to understand some papers by Mark van der Laan. He's a theoretical statistician at Berkeley working on problems overlap significantly with machine learning. One problem for me (besides ...
1
vote
0answers
22 views

Estimate E[x|A,B]: alternatives to bucketing for non-parametric estimation

I have a set of products. I would like to estimate Expected Value of items sold of the products wrt product price and age of the purchaser. One alternative is to assume a distribution and fit it. ...
0
votes
0answers
8 views

Relationships among different definitions of Sobolev spaces [migrated]

In Tsybakov's book(Page 51), Sobolev space (or Ellipsoid) for positive smoothness parameter $s$ is defined with sequential model, i.e. the series of the Fourier coefficients is finite. On the other ...
1
vote
0answers
11 views

Non-Parametric ANOVA: Repeated Measures with Replication

I'm struggling to find out which ANOVA test to use when you have a non-normal data set with repeated measures AND replication. I'm doing my data analysis in R and the friedman.test() method says it ...
1
vote
1answer
25 views

Chi-Square test: one-sided implementation?

I need to compare two samples and I would like to know if the distribution of the first stochastically dominates that of the second one. I'm not sure whether with the Chi-Square test I can verify ...
1
vote
1answer
37 views

Density Function Estimation

Given a sample of $n$ observations, which are assumed to be $i.i.d.$ and generated from a continuous probability law. Consider the question of estimating the density function $f(x)$. There are two ...
1
vote
0answers
5 views

What's the best way to demonstrate differences in these means/totals?

With some co-authors, I coded 200 video game reviews based on whether they mentioned any of nine different game elements. We now have the total number of times that each game element showed up in ...
1
vote
0answers
21 views

hypothesis test for discrete disributions

I am interested in the number of instances of feature $F$ among the members of two populations, $A_1$ and $A_2$. e.g. $F$ is a some special genetic element that may occur several times within a ...
3
votes
1answer
34 views

What is the optimal bandwidth in this problem

Suppose that I have some estimator $\hat f$ for the parameter $f$. I obtain the following convergence rate $$E\|\hat f - f\|^2 = O\left(\frac{1}{nh} + \frac{1}{nh^2} + \frac{1}{\sqrt{n}h} + ...
1
vote
0answers
15 views

Non-parametric Levene's test by Nordstokke and Zumbo

The example they mention is using a one-way ANOVA. What if I have two factors (3x11) and a dependent variable, can I do a two-way ANOVA to calculate the univariate levene's test? If so, how would I ...
3
votes
1answer
37 views

How to use the bunching estimator to estimate the elasticity of taxable income

Today my professor told us about a recent estimator that non-parametrically estimates the elasticity of taxable income. I understood why doing this gets around problems of previous studies that use ...
0
votes
0answers
18 views

Parameter estimation in lottery with some unique and some non-unique prizes

Let's consider this lottery: Each time a player plays he gets a prize. Either it's one of the normal prizes $a, b, c, ..$ or one of the unique prizes $A, B, C, ..$. Here "unique" prize is such that ...
0
votes
0answers
9 views

Bonferroni family of confidence intervals for fitted values from loess method

I did loess nonparametric regression using one predictor and I wanted to build confidence band around my fitted values. I saw the some posts(links below) for this topic and found that an approximation ...
7
votes
3answers
196 views

Why is the asymptotic relative efficiency of the Wilcoxon test $3/\pi$ compared to Student's t-test for normally distributed data?

It is well-known that the asymptotic relative efficiency (ARE) of the Wilcoxon signed rank test is $\frac{3}{\pi} \approx 0.955$ compared to Student's t-test, if the data are drawn from a normally ...
0
votes
0answers
26 views

Significant Friedman's Test, but non-significant post-hocs?

I have a sample of 25 in my within-subjects design, with my DV as error rates. The data contains a lot of zeros (i.e. no errors) and is highly negatively skewed, so I can't run the repeated-measure ...
0
votes
1answer
14 views

Dependent samples but unable to pair individual scores

We surveyed a group of students before an in-school year-long project and after, and want to see if the changes in scores are significant. Data doesn't meet assumptions of normality. Can we use the ...
2
votes
1answer
45 views

Paired samples or independent sample hypothesis test for two time periods

I want to know if employees in an organization are surveyed in 2013 and again in 2014 if the samples from the two time periods are considered related and dependent or independent if they are asked the ...
0
votes
0answers
16 views

Goodman Kruskal with ordinal and categorical variables

I have a data set of 100 people grading likelihood of future recall from highest to lowest for a set of memory targets. I am only interested in them making this judgment on unrecalled items, meaning ...
2
votes
1answer
57 views

Mann-Whitney U test when only summary data (mean, sd, sample size) are available

When raw data is not available, which normally occurs when we obtain our data from different scientific papers, how is it possible to perform a Mann-Whitney analysis to test whether two samples are ...
0
votes
0answers
27 views

What is the relationship between degrees of freedom and number of comparisons in Nemenyi's post hoc test?

I have some data consisting of non-normally distributed samples drawn from one of several different groups. A Kruskal-Wallis 1-way ANOVA rejects the null hypothesis that the medians of the groups are ...
1
vote
0answers
69 views

Estimating the 95th percentile of nitrate data using non-parametric Weibull method, with associated confidence intervals in R

I am currently involved in a project where I have to estimate the 95th percentile of nitrate data from different sample points, using the non-parametric Weibull method, to see if it breaches the ...
0
votes
1answer
54 views

multiple pair wise nonparametric comparisons

is it possible to perform multiple pair wise nonparametric comparisons (generally using a Mann–Whitney test) and adjust the significance level used for the decision criteria using a Bonferroni-type ...
1
vote
0answers
30 views

Multiple Comparisons testing and Dunn's Test

I have 8 different groups of individuals, each with a pre- and post-treatment measurement. The data are non-normal, so I used a Jonckheere-Terpstra test to test for an ordered alternative and ...
4
votes
0answers
32 views

Rank deficient bootstrap resamples

Despite years of stat courses I'm afraid I may still not completely understand bootstrapping. My question here relates to nonparametric boostrapping of regression models. As i understand it you draw ...
0
votes
0answers
30 views

lay stats question

I am comparing two prospective observational cohorts; the "study group" (n=30) has undergone previous intervention that "failed" and has undergone a 2nd intervention. This is being compared with a ...
1
vote
0answers
48 views

How to pool the interquartile range for an arbitrary distribution?

Let's say that I want to estimate the temperature in an annulus-like region. I split the annulus into four equal sectors (sectors 1,2,3, and 4) with $n_i$ being the number of measurements in sector ...
1
vote
1answer
53 views

Using nonparametric tests with small samples even when data are normaly distrubuted

Is it appropriate to use a nonparametric test with a small sample (e.g., less than 65 people) even when data are normally distributed? For example, using the Spearman correlation? Actually my data is ...
1
vote
1answer
38 views

nonparametric method to calculate the probability how alike two samples are

I have two samples with each couple of hunderd observations. I want to calculate a probabilty how much they look alike. I'm aware of tests like kolmogorov smirnov but I don't think I need this. I ...
0
votes
1answer
39 views

Likely mean of a multinomial distribution with dirichlet prior

I am working to create a Bayesian non-parametric estimate of the mean of a distribution given a distribution of observations. Ultimately I'd like to get to a credibility interval of the likely mean of ...
2
votes
1answer
29 views

What does “CRP is a marginalized version of PYP” mean?

I've been reading this phrase and I don't know what it means "CRP is a marginalized version of PYP". What are the parameters/latent-variables we are marginalizing out to drive CRP from PYP?
0
votes
0answers
41 views

How to perform Friedman's test with covariates

I have 500 participants who each completed 30 trials. In each trial they need to choose the correct image out of three (a three-alternative forced-choice task). Of these thirty trials, 10 trials ...
3
votes
1answer
61 views

Why are there two forms for the Mann-Whitney U test statistic?

I have encountered two forms for calculating the two-sample Mann-Whitney U test statistic, which are: $$U_1 = R_1 - \frac{n_1(n_1 + 1)}{2}$$ and $$U_1 = n_1n_2 + \frac{n_1(n_1+1)}{2} - R_1$$ where ...
1
vote
1answer
37 views

What is a one-way ANOVA on the ranks of the observations

While reading this link: http://www.statsdirect.com/help/Default.htm#nonparametric_methods/kruskal_wallis.htm, I came across the following: ...
0
votes
1answer
28 views

Reporting results of a Quade test

I have been scouring the internet to find out how to report the results of a Quade test (non-parametric equivalent of ANCOVA). Does someone have an example that explains this in plain terms? I'm not ...
1
vote
0answers
19 views

Is it okay to use lots of different statistical tests within the same dataset due to not fulfilling parametric assumptions?

I'm trying to reconcile the results of a study, and the fact that we need to do lots of different types of statistical tests to show a few conclusions in a paper. If the data were normally distributed ...
9
votes
1answer
118 views

Generalized additive models — who does research on them besides Simon Wood?

I use GAMs more and more. When I go to provide references for their various components (smoothing parameter selection, various spline bases, p-values of smooth terms), they are all from one ...
1
vote
0answers
21 views

Non-parametric effect size

Effect size, as estimated by Cohen's $\delta$, is: $d=\frac{\hat x_1 - \hat x_2}{\hat s_{pooled}}$ Since the numerator is an estimate of difference in location, and the denominator is an ...
1
vote
0answers
13 views

Risk in density estimation: grasping the definition

When generalizing estimators to an entire function what is the space in which we perform the integral to obtain the expected value (with respect to this function)? For example, when estimating ...
1
vote
2answers
118 views

Any other non-parametric alternative to Kruskal-Wallis?

It looks like Kruskal-Wallis is the standard nonparametric test for more than two groups. The problem is that it does not tell which groups are different, except that whether there exists significant ...
19
votes
6answers
599 views

A principled method for choosing between t test or non-parametric e.g. Wilcoxon in small samples

Certain hypothesis tests can be approached using Student's t-test, possibly using a Welch correction for unequal variances, or by a non-parametric test such as Wilcoxon paired signed rank test, the ...
2
votes
1answer
45 views

How to cope with non-normal ANOVA residuals?

My data follow this outline: ...
2
votes
1answer
44 views

non-parametric tests with significant difference in sample sizes

I have outcomes from a biological expt., where we have measured DNA lengths that has been affected. For condition1, I have 11 observations (length of DNA) and for condition2 I have approx. 116000. I ...
0
votes
0answers
12 views

What options do I have to analyze Non-parametric version of two-factor ANOVA with repetition?

Sorry, my previous question was not very clear I deleted it and am trying again. Question: What options do I have for two-factor non-parametric version of ANOVA with repetition? What is the ...
0
votes
0answers
38 views

How to make non-parametric distribution estimation with known, limited number of points of the CDF?

Is there any method to make non-parametric estimation of a cumulative distribution function (CDF) which actual points (not a sample) can be calculated numerically? I have a numerical method which can ...