STAT241 Regression & Modelling 1
| First Semester |
Regression and modeling, the “core techniques” of modern statistical analysis, appear regularly in the research journals of many fields ranging from the health sciences, nutrition, epidemiology, ecology, environmental science, zoology and botany to sociology, marketing, economics and finance. In many settings there is a response or outcome being researched and a number, often quite large, of potential causal factors. A regression analysis develops equations which assist in identifying important influences on an outcome measure. The regression procedures allow correction for potential confounding effects as well as suggesting hypotheses for future investigation with designed experiments. The paper uses the statistical packages SPSS 22 and R.
This paper is central for the advanced study of statistics and biostatistics. It is the first in a regression sequence which extends from second to fourth year in statistics. It is an excellent paper for a minor in statistics as well as the Diploma for Graduates endorsed in statistics. It is a prerequisite for the applied time series paper at third year level.
Paper details
Regression summarises (or models) complex data in a compact way and identifies factors which explain variability in an outcome measure. The data are frequently observational but some designed studies are also analysed. Economists use regression for forecasting, ecologists employ regression to study the effects of tourism on dolphin behaviour, sociologists build regression based causal models, a nutrition scientist models dietary factors which inhibit and enhance iron storage levels in newborn babies, epidemiologists investigate factors which influence cot death. These studies involve the systematic assessment of various exposure variables on an outcome of interest and make adjustments for covariates or confounders which affect both the outcome and possibly some exposure variables in the model. The methods developed are covered and involve simple linear, multiple linear and logistic regression procedures along with conditions required for the correct use of these methods.
Potential students
All students who intend to major in statistics or biostatistics should take this paper. It is a key paper for a minor in Statistics, for the DipGrad endorsed in Statistics or for a double major. The paper is useful as an advanced-service statistics paper for all students majoring in any of the subjects listed above. There is no mathematics prerequisite and the paper can be taken with a background of either STAT 110 or STAT 115. It is the first in a regression sequence that extends from second to fourth year in statistics. The paper is also useful as an advanced service statistics paper for all students majoring in any of the subjects listed above or for postgraduates in many other subjects. The paper uses the statistical package SPSS as well as R. It leads into the third year STAT 341, STAT 352 and STAT 380 as well as being helpful for Stat 342. These four papers can also be taken without mathematics prerequisites.
Prerequisites
STAT 110 or STAT 115
Main topics
- Simple linear regression
- Multiple linear regression
- Model building and model diagnostics
- Outliers and influential points
- Dummy variables, categorical predictors
- Factorial experiments and interactions
- Two factor unbalanced experiments
- Sequential data and interrupted time series
- Logistic regression for binary data
- Comparison of logistic models
- Confidence intervals for odds ratios
- Adjustments for covariates and confounders
- Binomial responses and overdispersion
- Multinomial logistic regression
Required text
None, course notes will be available for purchase at the University Print Shop.
Lecturers
Assoc Professor John Harraway, Room 238, Science III
Dr Tilman Davies, Room 516, Science III
Lectures
Tuesday ( SDAV2 ), Thursday ( RMOOT ) and Friday ( BURNS7 ) at 1pm.
We will sometimes alternate between all three and only two lectures a week; students will be informed well in advance. Total of 33 lectures.
Tutorials
One hour per week at times to be arranged. The tutorials begin in week 2 of the semester and are scheduled for 2pm on Monday, Wednesday or Thursday. They are all held in B21 Science III.
Internal Assessment
Nine or ten exercises contributing 50 marks.
A mid semester test in week 11 of the semester contributing 50 marks.
Exam format
A three-hour written examination worth 100 marks.
Final mark
Your final mark F in the paper will be calculated according to this formula:
F = max(E, (4E + A + T)/6)
where:
- E is the Exam mark
- A is the Assignments mark
- T is the Tests mark
and all quantities are expressed as percentages.
Students must abide by the University’s Academic Integrity Policy
Academic endeavours at the University of Otago are built upon an essential commitment to academic integrity.
The two most common forms of academic misconduct are plagiarism and unauthorised collaboration.
Academic misconduct: Plagiarism
Plagiarism is defined as:
- Copying or paraphrasing another person’s work and presenting it as your own.
- Being party to someone else’s plagiarism by letting them copy your work or helping them to copy the work of someone else without acknowledgement.
- Using your own work in another situation, such as for the assessment of a different paper or program, without indicating the source.
- Plagiarism can be unintentional or intentional. Even if it is unintentional, it is still considered to be plagiarism.
All students have a responsibility to be aware of acceptable academic practice in relation to the use of material prepared by others and are expected to take all steps reasonably necessary to ensure no breach of acceptable academic practice occurs. You should also be aware that plagiarism is easy to detect and the University has policies in place to deal with it.
Academic misconduct: Unauthorised Collaboration
Unauthorised Collaboration occurs when you work with, or share work with, others on an assessment which is designed as a task for individuals and in which individual answers are required. This form does not include assessment tasks where students are required or permitted to present their results as collaborative work. Nor does it preclude collaborative effort in research or study for assignments, tests or examinations; but unless it is explicitly stated otherwise, each student’s answers should be in their own words. If you are not sure if collaboration is allowed, check with your lecturer.
Iron Deficiency
Concern about iron deficiency in New Zealand infants and toddlers initiated a large-scale survey in the South Island by the University’s Department of Human Nutrition. The impact of breast feeding compared with use of cows milk is assessed after correcting for infections which can confound the result in both diet groups.
Stadium Support
Is there support or lack of support in a community to an expensive project using public funds? How do we carry out the survey and analyse the data? The profile of the Dunedin ratepayers who supported the building of the Dunedin covered stadium are identified from a postal survey which produced about 1800 respondents.
Preferred habitat
Habitat selection by wild animals is a major issue in environmental science, in particular in protecting the habitat of endangered species to ensure the survival of such a speciesThe New Zealand Hector’s dolphin is an endangered animal and a study is being carried out to establish factors which may determine a preferred habitat. These factors include water temperature, water clarity and water depth, as well as seasonal effects and prey abundance. Regression analysis identifies those factors that influence the dolphins’ choice of habitat. A set of data based on an investigation of 980 sites around the South Island, half the sites having dolphins present, is analysed in this paper. Some seasonal effects have been identified.
