Advanced Modelling
Code: 104865 ECTS Credits: 6| Degree | Type | Year |
|---|---|---|
| 2503852 Applied Statistics | OB | 3 |
Contact
- Name:
- Ferran Torres Benitez
- Email:
- ferran.torres@uab.cat
Teachers
- Jose Rios Guillermo
Teaching groups languages
You can view this information at the end of this document.
Prerequisites
Prior knowledge of sufficient knowledge in both theoretical statistics (linear models, statistical inference, and probability) and applied statistical software management. Internships can be followed with R, SAS or Stata.
A sufficient level of English is required to understand scientific articles to apply modeling knowledge.
Objectives and Contextualisation
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 (Competence) Assess the suitability of the models with the correct use and interpretation of indicators and graphs.
- CM09 (Competence) Assess the suitability of the models with the correct use and interpretation of indicators and graphs.
- CM10 (Competence) Modify the existing software if required by the statistic model, or create new software, if necessary.
- KM12 (Knowledge) Provide the experimental hypotheses of modelling, considering the technical and ethical implications involved.
- KM12 (Knowledge) Provide the experimental hypotheses of modelling, considering the technical and ethical implications involved.
- SM12 (Skill) Interpret the results obtained to formulate conclusions about the experimental hypotheses.
- SM14 (Skill) Use graphs to visualise the fit and suitability of the model.
Content
- Basic concepts in statistics applied to modeling
- Obtaining, supervising and preparing the data
- Effect measures and related models. Selection of models depending on the design
- Models used in studies with confounding factors and effect modifiers. Role of different (co) variables
- Application of multivariate and regression logistic regression models
- Propensity score and other alternatives for control of confounding factors
- Adjusted meta-analysis for individual data
- Adjusted repeated measurements with fixed and random effects
There will be practical examples in each block and students will have to deliver the practices done in groups
Activities and Methodology
| Title | Hours | ECTS | Learning Outcomes |
|---|---|---|---|
| Type: Directed | |||
| Practical lecture | 50 | 2 | |
| Theory 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.
Annotation: Within the schedule set by the centre or degree programme, 15 minutes of one class will be reserved for students to evaluate their lecturers and their courses or modules through questionnaires.
Assessment
Continous Assessment Activities
| Title | Weighting | Hours | ECTS | Learning Outcomes |
|---|---|---|---|---|
| Exam 1 | 15% | 6 | 0.24 | CM09, CM10, KM12, SM12, SM14 |
| Exam 2 | 15% | 6 | 0.24 | CM09, CM10, KM12, SM12, SM14 |
| On-site continuous assessment during classes | 15% | 4 | 0.16 | CM09, CM10, KM12, SM12, SM14 |
| Practical works | 45% | 30 | 1.2 | CM09, CM10, KM12, SM12, SM14 |
| Self-learning tests | 10% | 4 | 0.16 | CM09, CM10, KM12, SM12, 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
Bibliography
Faraway, J. (2006). Extending the Linear Model with R. Chapman & Hall.
Hosmer, D.W.; Lemeshow, S. & Sturdivant, R.X. (2013) Applied Logistic Regression. 3rd ed. Wiley.
Pinheiro JC & Bates D (2000) Mixed-Effects Models in S and S-PLUS. Springer.
T Hastie, R Tibshirani, J Friedman. (2009) The Elements of Statistical Learning. Data Mining, Inference and Prediction, Springer, New York.
Therneau T, Grambsch P. Modeling Survival Data: Extending the Cox Model (Statistics for Biology and Health). Springer-Verlag New York Inc.; Edición: 1st ed. 2000.
Venables, W. & Ripley, B. (2002). Modern Applied Statistics with S-PLUS. Springer
Verbeke G, Molenberghs G. Linear Mixed Models for longitudinal Data. New York: Springer-Verlag, 2000.
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).
Language list
| Name | Group | Language | Semester | Turn |
|---|---|---|---|---|
| (PLAB) Practical laboratories | 1 | Catalan | second semester | afternoon |
| (TE) Theory | 1 | Catalan | second semester | afternoon |