- Docente: Silvia Pacei
- Credits: 6
- SSD: SECS-S/03
- Language: Italian
- Teaching Mode: Traditional lectures
- Campus: Bologna
-
Corso:
Second cycle degree programme (LM) in
Statistics, Economics and Business (cod. 8876)
Also valid for Second cycle degree programme (LM) in Statistics, Economics and Business (cod. 8876)
Course contents
1. THE NON-SAMPLING ERROR
1.1. Sources of errors.
1.2. The “non-response” and its consequences.
1.3. Three types of non-response: a) errors in the list; b) total non-response; c) item non-response.
2. ERRORS IN THE LIST
2.1. Population, list and method of interview: main schemes for kind of list and method of interview.
2.2. Types of error in the lists: non-complete list; over-coverage in the list; double units in the list; list with grapes of units.
2.3. Methods to control for the errors in the list.
3. TOTAL NON-RESPONSE
3.1. Reasons for total non-response. Consequenses of total non-response.
3.2. How to prevent total non-responses or correct them during the survey.
3.3. Total non-response correct in the estimation phase: 1. Classes of adjustment for non-response; 2. Two-phase method; 3. Regression estimator; 4. Constrained-weighted estimator.
3.4. Non-response schemes and auxiliary variables selection.
4. ITEM NON-RESPONSE
4.1. Typologies of item non-response.
4.2. How to prevent the item non-responses.
4.3 Methods to deal with incomplete data: 1. Methods only based on respondents; 2. Weighting Methods; 3. Imputation Methods (Deductive Imputation; Imputation of the mean; Imputation from donator).
4.4. Imputation Methods for longitudinal surveys.
4.5. Comparative Analysis among imputation methods.
4.6. Multiple Imputation.
POINTS 5 AND 6 ARE EXCLUDED FROM THE "METODI DI CAMPIONAMENTO E INDAGINI CAMPIONARIE" PROGRAM
5. Examples of application in official surveys: Banca D’Italia sample survey on Italian households income; ISTAT sample survey on small and medium firms.
6. Introduction to the small area estimation problem.
SAS PROGRAMMING ON REAL OR SIMULATED DATASET.
Readings/Bibliography
Slides.
G. Nicolini; D. Marasini; G.E. Montanari; M. Pratesi; M.G. Ranalli; E. Rocco (2013). Metodi di stima in presenza di errori non campionari. Milano: Springer-Verlag Italia. Chapters 4, 5 and 6.
Teaching methods
Lectures and laboratory exercises using the SAS software.
Assessment methods
Laboratory exam with eligibility, preparatory to the oral exam with a mark, to be taken in the same exam call.
Office hours
See the website of Silvia Pacei