. - Quantitative Methods
PROFF. PAOLO SCKOKAI- CLAUDIA LANCIOTTI
Applied Agricultural and Food Economics
PROF.PAOLO SCKOKAI
COURSE AIMS
The course aims to introduce students to some basic econometric tools applied
to food and agricultural data. Special attention will be given to those models that
can be applied in a business environment.
COURSE CONTENT
The multiple regression model. Review of the multiple regression
model. The use of dummy variables. F-tests on model specification.
Heteroscedasticity and serial correlation. Definition, test and
correction for heteroscedasticity. Definition, test and correction for
serial correlation.
Forecasting. Use of the regression model for forecasting. Forecast
error, forecast error variance and confidence intervals. Forecasting
with serially correlated errors.
Models of qualitative choice. Definitions: binary and multiple choice
models. Binary Linear Probability Model, Probit Model and Logit
Model.
Panel data models. OLS estimation on panel data. Fixed-effects
models: definition and estimation problems.
Instrumental Variable estimation. Correlation between
explanatory variables and error term. IV estimation, endogeneity and
2SLS estimation.
Estimation of systems of equations. Simultaneity, OLS estimation
and the identification problem. Seemingly unrelated regressions
(SUR) and Three-stage least squares (3SLS).
Tutorial sessions
READING LIST
Selected readings from the following textbook
ECTS
1.0
0.5
0.5
1.0
0.5
0.5
1.0
1.0
R.S PINDYCK-D.L. RUBINFELD, Econometric Models and Economic Forecasts, 4a ed., McGraw-Hill,
1998.
Further readings on specific topics will be provided by the instructor.
TEACHING METHOD
The course consists of five credits of lectures and one credit of tutorial computer
sessions.
ASSESSMENT METHOD
The exam is structured in two parts: one final written exam (75% of the final grade) and
three homeworks carried out in small groups (25% of the final grade).
The final written exam lasts 120 minutes and it is structured with open questions and
exercises concerning applications of econometric models to agricultural and food economic
data. The score attached to each question may change depending on the test. The assessment
is intended to provide a sufficiently precise measure of the student’s learning and to offer to
the instructor a grasp of the student’s reasoning skills and abilities to use econometric tools.
The three homeworks will refer to the content of the tutorial computer sessions and will
ask the students to carry out econometric analysis of agricultural and food economic data
using a specific econometric software.
NOTES
Further information can be found on the instructor's webpage or on the Faculty notice
board.
Prof. Paolo Sckokai is available to meet with studentes after class in the SMEA offices.
Applied Statistics for the Agri-Food System
PROFESSOR CLAUDIA LANCIOTTI
COURSE AIMS
The course develops the student’s ability to use descriptive and inferential
statistics in data analysis, elaboration of hypotheses and in reaching conclusions in
the food sector applied research.
COURSE CONTENT
Descriptive statistics.
Presentation and summary of univariate data, presentation and
summary of bivariate data, graphical depiction. Measures of
central tendency and measures of variability, covariance and
correlation, concentration analysis. Index numbers.
Probability and probability distribution.
Theories of probability, theorems. Binominal distribution,
uniform distribution, normal distribution and central limit
theorem.
Sampling and sampling distributions.
Techniques, sample size, sampling and non-sampling errors.
Random variables, estimators and their properties.
STATISTICAL INFERENCE
Point estimate and interval estimate of population parameters, t,
2 and F distributions. Hypothesis tests for single populations.
Hypothesis tests about the comparison of two independent or
related populations. Chi-squared goodness-of-fit test and chisquared test of independence.
Analysis of variance.
One-way ANOVA. Tukey’s and Tukey-Kramer’s multiple
comparison tests. Two-way ANOVA.
Simple linear regression analysis.
Least squares estimates. Test of the assumptions and residual
analysis. Hypothesis tests for the regression coefficient and
testing the overall model. Confidence intervals for y|x and
prediction intervals for y.
Multiple linear regression analysis.
Hypothesis tests for the regression coefficients and partial F test,
adjusted R2, residual analysis. Non-linear regression models.
Dummy variables. Multicollinearity and model-building
procedures.
Tutorials.
CFU
0.5
0.5
0.5
1.5
0.5
1.0
0.5
1.0
READING LIST
K. BLACK, Business Statistics for Contemporary Decision Making, 5e, John Wiley & Sons, USA,
2007.
Slides and classnotes.
TEACHING METHOD
Lectures, for five credits, and tutorials, for one credits, will be held in the computer lab.
ASSESSMENT METHOD
The exam is written and oral.
The written part consists of two midterm tests.
The first midterm test, set up after the first two weeks of lectures, aims to assess the
capacity of the student to summarize, present, understand relations in a data set using a
spreadsheet. The test lasts 120 minutes and its evaluation accounts for 30% of the final
grade.
The second midterm test takes place after 80% of the course and its purpose is to assess
learning of the use of probabilities and the major statistical tools (estimations and statistical
tests) in problems with limited information. The test lasts 120 minutes and its evaluation
accounts for 30% of the final grade.
The oral part is the final one and concerns a discussion and presentation of an individual
homework. Its purpose is to assess the mastering of tools related to a model building and the
skills of the student in presenting and commenting a problem in the real life. The final part
of the exam account for 40% of the final grade.
NOTES
Professor Claudia Lanciotti is availlable to meet with students by appointment.