Statistical Analysis Stack Exchange
Statistical Analysis Stack Exchange

Questions tagged [satterthwaite]

Refers to statistician Franklin E. Satterthwaite and his associated work.

This tag is appropriate for questions regarding the theory or implementation of statistical procedures related to Franklin E. Satterthwaite's work.

Most are probably familiar with his suggested formula for the degrees of freedom for the unequal variance two-sample t-test, which was also discovered seemingly independently around the same time by B.L. Welch: Welch, B.L. (1938). The significance of the difference between two means when the population variances are unequal. Biometrika 29, 3/4, 350–62.

One of Satterthwaite's most famous works regarding t-tests for means is: Satterthwaite, F.E. (1946). An approximate distribution of estimates of variance components. Biometrics Bulletin 2, 6, 110–114.

7 questions
11
votes
3 answers

Welch test seems to perform much worse than equal variance t-test

The SciPy function in Python, ttest_ind() by default works with the $t$-test that assumes equal variances. There is a parameter, equal_var = False that switches it to the Welch test where equal variances in the two samples are not assumed. This…
ryu576
  • 2,220
  • 1
  • 16
  • 25
8
votes
1 answer

Is there ever a statistical reason NOT to use Satterthwaite's method to account for unequal variances?

Two related questions: 1) From what I've read, using Satterthwaite degrees of freedom does not assume equal variances, which allows it to be used on datasets that would be inappropriate for, say, a vanilla ANOVA. However, from a conceptual…
amateur3057
  • 239
  • 1
  • 5
4
votes
1 answer

R's t.test() unequal variance degrees of freedom

Can anyone explain why, for an independent two-sample t-test assuming unequal variances, R's t.test function uses the label "Welch Two Sample t-test," when it appears to use Satterthwaite's 1946 formula for the degrees of freedom (d.f.)…
Meg
  • 1,633
  • 4
  • 18
  • 28
3
votes
0 answers

The "generalized" Satterthwaite method?

Consider the one-way ANOVA with random effect. library(AOV1R) # a package of mine set.seed(666) I=3; J=4 $ number of groups and number of replicates dat <- simAV1R(I, J, mu=0, sigmab=2, sigmaw=3) fit <- aov1r(y ~ group, data=dat) ssb <- fit[["Sums…
Stéphane Laurent
  • 17,425
  • 5
  • 59
  • 101
1
vote
1 answer

Non significant intercept but significant coefficients in mixed effect modelling

I am using mixed effect models to predict a time series of data. I am using lmertest() in R to overload lmer() to gain p values via Satterthwaite approximation. The general model for each formula in R syntax is: Dependent Variable ~ time^2 + time +…
Alesi Rowland
  • 197
  • 8
1
vote
0 answers

specification of random slopes for interaction terms in lmer

I have a linear mixed model specified as follows model1<-lmer(y~A*B+(1|region/plot)+(A*B|region),data=mydata) where A is a categorical variable with two levels and B is a continuous variable. The dataset contains 80 observations from 40 plots and 3…
Jan
  • 65
  • 1
  • 8
1
vote
0 answers

Mixed models with Satterthwaite and AR(1) together

I have a question which is half theoretical and half practical. I am working on longitudinal datasets with small sample sizes, which i analyze with mixed models. Now, i am aware of the possible corrections in such cases, Satterthwaite and…
GiannisZ
  • 195
  • 1
  • 12