Experimental design and analysis
Master’s study programme Viticulture and Enology
Objectives and competences
Proper experimental design is a prerequisite to the efficient and cost effective resolution of comparative quantitative research questions. This subject introduces experimental design and analysis through the use of examples and by the study of underlying linear model. Use of appropriate computer packages allows testing of assumptions and investigation of advanced topics. Extensions of the basic methodology are explored. Students will gain experience in evaluating and presenting results in the form of a standard statistical report as used in professional practice. The experiences will be used in the evaluation of the results of different experiments (scientific and professional) in the field.
Prerequisites
Admission to the first and second year of study program.
Content
- Principles of experimental design.
- Simple linear regression: review and extension.
- Polynomial and multiple linear regression.
- Principles of experimental design; including completely randomized designs, randomized block designs and Latin square designs.
- Multiple comparison methods.
- Diagnostic checking of the basic model.
- The analysis of experiments with factorial treatment structures.
- Analysis of variance.
- Analysis of covariance.
- Statistical software.
Intended learning outcomes
Knowledge and understanding:
• be able to identify situations for which a linear model is appropriate;
* be able to design, and check the design of standard experiments reflective of professional practice;
* be able to perform the analysis of such experiments using appropriate statistical protocols;
* be able to communicate results and justify interpretations in terms of the original problem with logic and precision;
* be able to evaluate and present results, with an integrated understanding of the underlying theory, in the form of a standard statistical report;
* be able to critically test any assumptions underpinning the use of the linear model;
* be able to perform design and analysis tasks on recognized software platforms
Readings
- Gary W. Oehlert, A first Course in Design and Analysis of Experiments, 2010, E-version
- Howard J. Seltman, Experimental Design and Analysis,2015, E-version
- Adamič Š. (2005).Temelji biostatistike, Medicinska fakulteta Ljubljana, 1995.
- Košmelj, K. (2007). Uporabna statistika. Ljubljana: Univerza v Ljubljani, Biotehniška fakulteta E-version
- A. Vadnjal, Elementarni uvod v verjetnostni račun, DZS, Ljubljana, 1979. Catalogue
- Internet resources for learning statistics
- Moodle classroom
- Slides from lectures
Assessment
Seminar work and homeworks (30%), written exam (70%)
Lecturer's references
Doc. dr. Ahmad Hosseini je doktoriral na Univerzi Saarland (Nemčija) in Univerzi Sabanci (Turčija) na področju operacijskih raziskav kot področja industrijskega inženirstva pri razporejanju omejenih virov (z manjšim poudarkom na računalništvu). Po doktorskem študiju je sodeloval pri dveh nacionalnih projektih s področja industrijske optimizacije, logistike, sistemskega inženirstva in gozdnega inženirstva (kot podoktorski sodelavec) na različnih oddelkih Univerze Umeå (Švedska), kjer je bil tudi predavatelj/nadzornik na oddelkih za industrijsko inženirstvo, management in matematiko.
Zanimajo ga predvsem interdisciplinarne raziskave na presečišču industrijskega inženirstva, operacijskih raziskav (OR), računalništva, uporabne matematike in podatkovne analitike, s poudarkom na teoriji, načrtovanju in izvajanju. Zlasti ga zanimajo modeliranje in reševanje kombinatoričnih in industrijskih optimizacijskih problemov na področju oskrbovalnih verig, sistemskega inženirstva, proizvodnje, logistike in distribucije, gozdnega inženirstva/načrtovanja ter transporta. Njegove raziskave temeljijo predvsem na tehnikah matematičnega programiranja, algoritmih za optimizacijo omrežij, hevristiki in kombinaciji (meta)hevristike z natančnimi, aproksimacijskimi ali stohastičnimi tehnikami optimizacije.
Trenutno je docent na Poslovno-tehniški fakulteti (PTF) ter Fakulteti za vinogradništvo in vinarstvo (FVV) Univerze v Novi Gorici, povezan pa je tudi s Centrom za informacijske tehnologije in uporabno matematiko.
Izbrane objave
- ČESNIK, Urban, MARTELANC, Mitja, OVSTHUS, Ingunn, RADOVANOVIĆ, Tatjana, HOSSEINI, Ahmad, MOZETIČ VODOPIVEC, Branka, BUTINAR, Lorena.
Functional characterization of Saccharomyces yeasts from cider produced in Hardanger.
Fermentation. 2023, vol. 9, issue 9, [article no.] 824, str. 1-27. ISSN 2311-5637.
DOI: 10.3390/fermentation9090824. [COBISS.SI-ID 164729091], [JCR, SNIP, Scopus] - HOSSEINI, Ahmad, WADBRO, Eddie, NGOC DO, Dung, LINDROOS, Ola
A scenario-based metaheuristic and optimization framework for cost-effective machine-trail network design in forestry. Computers and electronics in agriculture. [Print ed.]. Sep. 2023, vol. 212, [article no.] 108059, str. 1-13, ilustr. ISSN 0168-1699. DOI: 10.1016/j.compag.2023.108059. [COBISS.SI-ID 159537155], [JCR, SNIP, WoS, Scopus] - HOSSEINI, Ahmad, WADBRO, Eddie.
A hybrid greedy randomized heuristic for designing uncertain transport network layout. Expert systems with applications. [Print ed.]. Mar. 2022, vol. 190, [article no.] 116151, str. 1-10, ilustr. ISSN 0957-4174.
DOI: 10.1016/j.eswa.2021.116151. [COBISS.SI-ID 141605891], [JCR, SNIP, WoS, Scopus] - HOSSEINI, Ahmad, PISHVAEE, Mir Saman.
Capacity reliability under uncertainty in transportation networks: an optimization framework and stability assessment methodology. Fuzzy optimization and decision making. Sep. 2022, vol. 21, iss. 3, str. 479–512, ilustr. ISSN 1568-4539. DOI: 10.1007/s10700-021-09374-9. [COBISS.SI-ID 141610755], [JCR, SNIP, WoS, Scopus] - NAKHAEI, Niknaz, EBRAHIMI, Morteza, HOSSEINI, Ahmad.
A solution technique to cascading link failure prediction. Knowledge-based systems. [Print ed.]. Dec. 2022, vol. 258, [article no.] 109920, str. 1-14, ilustr. ISSN 0950-7051.
DOI: 10.1016/j.knosys.2022.109920. [COBISS.SI-ID 141591555], [JCR, SNIP, WoS, Scopus] - HOSSEINI, Ahmad, PISHVAEE, Mir Saman.
Extended computational formulations for tolerance-based sensitivity analysis of uncertain transportation networks.
Expert systems with applications. [Print ed.]. Nov. 2021, vol. 183, [article no.] 115252, str. 1-19, ilustr. ISSN 0957-4174. DOI: 10.1016/j.eswa.2021.115252. [COBISS.SI-ID 141608451], [JCR, SNIP, WoS, Scopus] - HOSSEINI, Ahmad, LINDROOS, Ola, WADBRO, Eddie.
A holistic optimization framework for forest machine trail network design accounting for multiple objectives and machines. Canadian journal of forest research. 2019, vol. 49, no. 2, str. 111-120, ilustr. ISSN 0045-5067.
DOI: 10.1139/cjfr-2018-0258. [COBISS.SI-ID 141605635], [JCR, SNIP, WoS, Scopus