
Advanced Modelling
Code: 104865Credits: 6
| Degree programme | Type | Course |
|---|---|---|
| Applied Statistics | OB | 3 |
Contact lecturer
- Name :
- Ferran Torres Benitez
- Email :
- ferran.torres@uab.cat
Teaching staff
- José Rios Guillermo
- Andrea Cadiz Giraldez
Group languages
You can consult this information at the end of the document.
Prerequisites
Students are expected to have prior knowledge of both theoretical statistics (linear models, statistical inference, and probability theory) and the practical use of statistical software. Practical sessions will be conducted using R; however, they may alternatively be completed using SAS or Stata. In the latter case, this must be agreed with the course coordinator before the practical sessions begin.
Students are also expected to have a sufficient command of English to read and understand scientific papers and to apply the modelling concepts derived from them.
Objectives
Learn different modeling strategies for data analysis, both in terms of the theoretical aspect and its applications. Provide applied knowledge in terms of design, organization, implementation, supervision, analysis, interpretation and dissemination of results.
The general objectives of the subject are:
1. Know the basics for the application of different models
2. Understand criteria for selecting variables based on objectives
3. Acquire knowledge about the interpretation and implications of different models
4. Acquire and apply programming knowledge
Learning outcomes
- CM09 (Assess the suitability of the models with the correct use and interpretation of indicators and graphs.) Assess the suitability of the models with the correct use and interpretation of indicators and graphs.
- KM12 (Provide the experimental hypotheses of modelling, considering the technical and ethical implications involved.) Provide the experimental hypotheses of modelling, considering the technical and ethical implications involved.
- SM12 (Interpret the results obtained to formulate conclusions about the experimental hypotheses.) Interpret the results obtained to formulate conclusions about the experimental hypotheses.
- SM13 (Compare the degree of adjustment between diverse statistical models.) Compare the degree of adjustment between diverse statistical models.
- SM14 (Use graphs to visualise the fit and suitability of the model.) Use graphs to visualise the fit and suitability of the model.
Contents
- Fundamental concepts in statistics applied to modeling.
- Data collection, supervision, and preparation processes for analysis.
- Measures of effect and related models. Model selection according to study design.
- Models used in studies with confounding factors and effect modifiers. Role of different (co)variables.
- Application and variants of multivariable logistic regression and survival models.
- Propensity score and other alternatives for confounding control.
- Individual participant data meta-analysis.
- Analysis of repeated measures using fixed and random effects models.
- Evaluation of gender perspective and of vulnerable groups.
There will be practical examples in each unit, and students will be required to submit group assignments.
Learning activities and methodology
| Title | Hours | ECTS | Learning outcomes |
|---|---|---|---|
| Theory lecture | 50 | 2 | |
| Practical lecture | 50 | 2 |
Supervised activities
- Theoretical classes (TE). Each thematic block will begin with one or more face-to-face theory classes where the teacher will explain the key concepts, encourage interaction and discussion of doubts, and give guidelines for monitoring and preparation of complementary activities..
The support teaching material will contain the essential contents of the theoretical classes, will be available in advance on the Virtual Campus of the subject, and it is recommended that students have it available during the class (computer, tablet or paper format) to facilitate its monitoring.
- Laboratory Practices (PLAB). Practices related to the theoretical concepts will be carried out. Work will be done to expand and consolidate previous scientific and technical knowledge, and scientific articles will be used to encourage discussion.
Autonomous activities
- Self-study tests with feedback will be provided, using the questionnaire utilities of the Moodle classroom of the virtual campus of the subject, to facilitate the review of the subject synchronized with the teaching of the syllabus.
- Group work. There will be several teams works in which students will try to apply their knowledge to a real situation under the supervision of the teacher. Problems will be solved by consulting different sources and using statistical software. The student's capacity for analysis, reasoning and expertise in solving problems related to the professional field will be promoted.
- Personal study. Although the subject is eminently focused on the practical implementation of knowledge in advanced modelling, there will be a minimum individual effort in order to assimilate the theoretical classes.
Tutorials and personal attention to students
Students are expected to attend classes and consult doubts by actively participating in their discussion. However, students can consult with the professors using the foro of the virtual campus and the e-mails indicated in the teaching staff.
Assessment
Continuous assessment activities
| Title | Weight | Hours | ECTS | Learning outcomes |
|---|---|---|---|---|
| On-site continuous assessment during classes | 15% | 4 | 0.16 | CM09, KM12, SM12, SM13, SM14 |
| Exam 1 | 15% | 6 | 0.24 | CM09, KM12, SM12, SM13, SM14 |
| Practical works | 45% | 30 | 1.2 | CM09, KM12, SM12, SM13, SM14 |
| Exam 2 | 15% | 6 | 0.24 | CM09, KM12, SM12, SM13, SM14 |
| Self-learning tests | 10% | 4 | 0.16 | CM09, KM12, SM12, SM13, SM14 |
If the criteria for averaging are met, then the final mark for the course will be calculated using the weightings described in this section. Otherwise, it will be necessary to recover the affected activities in order to make up the average. A minimum mark of 5 out of 10 points is required to pass the course.
To assess the degree of achievement of the competences, the following instruments and weightings will be used:
Exams
There will be two partial exams with a weighting of 15% each, in which students will have to answer questions on theoretical and applied concepts. The minimum mark for weighting is 3 out of 10.
These activities are compulsory. In order to have access to the recovery it is necessary to have done 80% of the evaluable activities and to have taken the 2 mid-term exams.
Practical work
These activities are compulsory and it is necessary to have at least a mark of 4 out of 10 in each of them, otherwise it will be necessary to recover the affected activities. Practical work is worth 45% of the overall mark for the course.
Deliveries after the deadline:
- The late delivery of the practices will imply a penalty of 20% of the obtained mark.
These activities are compulsory and recoverable.
Self-study activities
They will have a weight of 10% provided that at least 80% of the activities have been carried out, otherwise the mark for this part will be a zero. There is no minimum grade for these activities.
Deliveries after the deadline:
- The delivery of these activities late and up to 48 hours after the deadline, will imply a penalty of 20% on the grade obtained.
- The late delivery of activities after this 48-hour marginwillmean that they will be counted as not having beencompleted for the evaluation.
These activities are not mandatory, but they are not recoverable either.
Continuous training and evaluation
It is reminded that the evaluation will be made according to the contents commented by the teacher in class, and that, therefore, attendance in person is highly recommended since not all the information will be accessible on the virtual campus.
In addition, during the course there will be a continuous assessment and it will be necessary to have participated in 80% of the assessment activities for them to be weighted at 15%, otherwise the mark for this part will be a zero. Standard teaching innovation tools will be used to control class participation. There is no minimum mark for these activities.
These activities are not mandatory, but they are not recoverable either.
Summary of criteria and weights for the evaluation of the subject
|
|
Participation1 | minimum Participation2 | minimum Mark3 | Exercise recoverable4 | Weights5 |
| Test 1stpartial | Compulsory | 100% | 3 | Compulsory | 15% |
| Test 2ndpartial | Compulsory | 100% | 3 | Compulsory | 15% |
| Practical work | Compulsory | 100% | 4 | Compulsory | 45% |
| Self- study | Volunteer | ≥80% | NA | Unrecoverable | 10% |
| Continued appraisal | Volunteer | ≥80% | NA | Unrecoverable | 15% |
NA: Not applicable
1: Compulsory participation implies that non-participation will have to be recovered in order to be weighted, and if it is not done, it will not be possible to average, and therefore the subject will not be approved either. Voluntary participation implies that it is not compulsory but that it cannotberecovered later
2: Value of minimum participation to weight, otherwise the activities will count as 0
3: Minimum mark of 10 points to be weighted with the rest, if the minimum is not reached, the specific activity will have to be recovered, regardless of the rest of the marks of the same type
4: When the activity is recoverable, it must be recovered if the minimum mark is not obtained. In case of non-recoverable activity, the mark cannot be recovered, and therefore it will be weighted to the final mark, even if it is 0 or less than any threshold
5: Weight value if the previous criteria are met
Use of Artificial Intelligence (AI)
In this course, the use of Artificial Intelligence (AI) technologies is permitted as an integral part of coursework, provided that the final submission demonstrates a substantial contribution from the student in terms of analysis and critical reflection. Students must clearly identify which parts of their work have been generated using AI, specify the tools employed, and include a critical reflection on how these tools have influenced both the process and the final outcome of the assignment. Failure to disclose the use of AI will be regarded as a breach of academic integrity and may result in a penalty to the assessment grade or, in serious cases, more severe disciplinary sanctions.
Irregularities in Assessment Activities
Any irregularity committed during an assessment activity (including academic fraud, plagiarism, or the improper use of AI, unless such use is expressly authorised in the course guide) that may lead to a significant alteration of the assessment result will result in that assessment being awarded a mark of 0 (Fail). Where the course guide establishes that obtaining a minimum mark in that assessment activity is an essential requirement to pass the course, or where multiple irregularities are committed in the assessment activities of the same course, the final grade for the course will be 0 (Fail). In addition, disciplinary proceedings may be initiated against any student who commits any of these irregularities.
Bibliography
- Faraway, Julian James. (2016). Extending the linear model with R : generalized linear, Fara
mixed effects and nonparametric regression models. (2nd ed.) CRC Press Taylor & Francis
Group
Available in print at the libraryDisponible
- Hastie, Trevor & Friedman, Jerome & Tibshirani, Robert. (2009). The Elements of statistical Hast
learning : data mining, inference, and prediction. (2nd ed.) Springer
Available in print at the libraryDisponible
- Hastie, Trevor & Friedman, Jerome & Tibshirani, Robert. (2009). The Elements of statistical Hast
learning : data mining, inference, and prediction. (2nd ed.) Springer
Available onlineDisponible
- Hosmer, David W. & Sturdivant, Rodney X. & Lemeshow, Stanley. (2013). Applied logistic Hos
regression. (3rd ed.) Wiley
Available onlineDisponible
- Pinheiro, José C. & Bates, Douglas M. (2000). Mixed-effects models in S and S-PLUS. Pinh
Springer
Available onlineDisponible
- Therneau, Terry M. & Grambsch, Patricia M. (2000). Modeling survival data : extending the Ther
Cox model. Springer
Available in print at the libraryDisponible
- Venables, W. N. & Ripley, Brian D. (2002). Modern applied statistics with S. (4th ed.) Ven
Springer-Verlag
Available in print at the libraryDisponible
- Verbeke, Geert & Molenberghs, Geert. (2000). Linear mixed models for longitudinal data. Verb
(1st ed.) Springer
Available onlineDisponible
- Verbeke, Geert & Molenberghs, Geert. (2000). Linear mixed models for longitudinal data. Verb
Springer-Verlag
Available in print at the libraryDisponible
Software
SAS version 9.4 software (© SAS Institute Inc., Cary, NC, USA)
STATA (© Stata Corporation, College Station, TX, USA) and
R (© 2010 R free software foundation: http://www.r-project.org).
Course groups and languages
The information provided is provisional until November 30. After this date, you will be able to consult the language of each group through this link. To access the information, you will need to enter the course CODE
| Type of teaching | Group | Language | Semester | Shift |
|---|---|---|---|---|
| (TE) Theory | 1 | Catalan | second semester | afternoon |
| (PLAB) Practical laboratories | 1 | Catalan | second semester | afternoon |