Scribd
Upload
18 views2 pages

Sheet 4

This document outlines exercise sheets for a Statistics in Toxicology II course at TU Dortmund University for the winter semester 2024/2025. It includes exercises on the insert-and-absorb algorithm, Dunnett and Williams tests, variance stabilizing transformations, and multiple comparison procedures, each with specific tasks and points assigned. Students are required to submit their completed exercises and R code by December 3, 2024.

Uploaded by

Mehedi Hasan Ridoy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
18 views2 pages

Sheet 4

This document outlines exercise sheets for a Statistics in Toxicology II course at TU Dortmund University for the winter semester 2024/2025. It includes exercises on the insert-and-absorb algorithm, Dunnett and Williams tests, variance stabilizing transformations, and multiple comparison procedures, each with specific tasks and points assigned. Students are required to submit their completed exercises and R code by December 3, 2024.

Uploaded by

Mehedi Hasan Ridoy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

TU Dortmund University Winter semester 2024/2025

Department of Statistics 21. November 2024


JProf. Dr. Kirsten Schorning
M. Sc. Leonie Schürmeyer

Statistics in Toxicology II (Testing)


Exercise sheet 4

Exercise 10: Insert-and-absorb algorithm (10 points (6+4))


(a) In a toxicological study with six treatments T1,. . .,T6 (corresponding to an increasing
dose), in pairwise comparisons the following differences are significant: T1 vs. T3, T2 vs.
T4, T3 vs. T4, T3 vs. T5 and T3 vs. T6. Perform the insert-and-absorb algorithm by hand
to derive the corresponding letter display.

(b) Sketch a possible profile that fits the situation from (a). For all doses, draw a boxplot and
specify a sample size. Justify your plot. Note that this exercise can also be solved by hand.

Exercise 11: Dunnett and Williams test (6 points (1+1+1+3))


Consider a toxicological experiment with a control and k = 8 increasing doses. How many rows
(contrasts) does the contrast matrix have for the following tests? Specify how you have calculated
these numbers.

(a) Dunnett test

(b) Williams test

(c) Combination of Dunnett and Williams test

(d) Combination of Dunnett and Williams test with Downturn effects

Exercise 12: Variance stabilizing transformation (14 points (3+3+3+5))


Let Xu be a family of random variables with E(Xu ) = u and Var(Xu ) = v(u) (see chapter 2.4,
slide 63-65). Then the variance stabilizing transformation f is defined by
Z x
1
f (x) = p du.
v(u)

(a) Calculate the variance stabilizing transformation for v(u) = (u + 7)−4 .

(b) For the Poisson distribution X ∼ Poi(λ) with parameter λ, λ ∈ R≥0 show that the
√
corresponding variance stabilizing transformation is f (u) ∝ u.

(c) For the Exponential distribution X ∼ Exp(λ) with parameter λ, λ ∈ R≥0 show that the
corresponding variance stabilizing transformation is f (u) ∝ log u.

(d) For the Binomial distribution X ∼ Bin(n, p) with parameters n and p, p ∈ (0, 1) show that
√
the corresponding variance stabilizing transformation is f (u) ∝ arcsin ( u). Consider the
random variable X̄ = X/n.
d √ 1
Hint: The derivative of the arcsine function is given by dx arcsin (x) = 1−x2
.
Exercise 13: Multiple comparison procedures (10 points (2+2+2+4))
Load the file data_gradedsheet.RData. It contains simulated response values for a toxicological
experiment measured for six increasing doses, in six replicates each. It is assumed that the first
dose corresponds to the control.

(a) Display the data via boxplots and describe the overall shape.

(b) Making use of the functions glht and cld from the package multcomp, calculate the
compact letter display (cld).

(c) Consider the following compact letter display for T1 to T6, respectively:
a, ab, abc, bcd, cd, d. Is the result of (b) equivalent to this? Justify your choice.

(d) Perform three types of multiple testing procedures, using the function glht from the
package multcomp: Dunnett, Williams, and extension of Dunnett and Williams including
downturn effects. Plot two-sided confidence intervals of the results. Compare and interpret
the results.

Submit the processed exercise sheets (PDF) and the R code (.R or .Rmd) via Moodle until
Wednesday, 03. December 2024, 8.30. Please note the information listed in the Moodle on the
submission formalities for this course.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy