Review of statistical inference methods; Comparison of two groups;
Association between categorical variables; Correlation; Simple and multiple linear regression;
ANOVA: An introduction; ANCOVA: An introduction;
Model Building with multiple regression and Regression diagnostics;
Logistic regression models for binary response variables; Introduction to
causal inference and evaluation methods.
Implementation of the studied methods with the software STATA.
P. Newbold, W.L. Carlson, B. Thorne (2013), Statistics for Business and Economics 8th edition,
Pearson.
As an alternative to Newbold et al:
Agresti Alan, Finlay Barbara. (2015) Statistical Methods for the Social
Sciences (4th edition) Pearson Prentice Hall.
Ulrich Kohler and Frauke Kreuter (2013), Data Analysis Using Stata, Third edition, Stata Press.
Additional useful material:
On-line material about STATA:
https://stats.idre.ucla.edu/stata/
It is necessary to get the STATA software.
Learning Objectives
Students will develop expertise to use and interpret statistical models for
continuous and binary response variables and the ability to employ and
interpret appropriate statistical methods for causal analyses with the software STATA.
Prerequisites
Basic elements of descriptive and inferential statistics
Teaching Methods
Lectures, sessions of exercises and labs
Further information
Additional teaching materials will be provided during the course through the e-learning platform
Type of Assessment
Computer lab exam
Course program
Review of descriptive statistics: Variables and their measurement; Measures of centrality and dispersion; Percentages, percentage points and relative changes.
Review of statistical inference methods: Statistics and sampling distributions; Point estimation;
Interval estimation; Statistical hypothesis tests.
Comparison of two Groups: Introduction; Comparing two proportions;
Comparing two means; Comparing means with dependent samples.
Analyzing Association Between Categorical Variables: Joint, marginal and
conditional frequency distributions; Statistical dependence and independence; Chi-squared test of independence; Residuals; Association
measures (difference between two proportions; relative risk, odds ratio).
Correlation:
coefficient;
Linear dependence; Covariance; Linear correlation
Inference for the linear correlation coefficient.
Linear Regression: Model assumptions; Interpretation of the model parameters; Least square estimation of the regression parameters; Fitted
values and residuals; Interpolation and extrapolation; Decomposition of
the sum of squares; R-squared; Inference in the regression model
(tests and confidence intervals for the slope; confidence intervals for
expecteed values and predicted values).
Introduction to multivariate relationships.
Multiple Regression: Model assumptions; Interpretation of the model
parameters; Fitted values and residuals; R-squared andMultiple correlation; Inference in multiple regression models (tests and
confidence intervals for multiple regression coefficients); Interaction
between predictors in their effects; Comparing regression models.
Introduction to analysis of Variance (ANOVA): Comparing several means: One-way
analysis of variance and F-test; Multiple comparisons of means;
Performing ANOVA by regression modeling; Two-way analysis of variance;
Comparing regression models.
Introduction to analysis of Co-Variance (ANCOVA): Regression models with quantitative
and categorical predictors; Interaction between quantitative and
categorical predictors; Inference for regression with quantitative and
categorical predictors; Comparing regression models.
Introduction to model selection procedures; Regression diagnostics;
Multicollinearity.
Logistic Regression for binary response variables:
Introduction to generalized linear models; Logistic regression; Multiple logistic
regression; Inference for logistic regression models; Comparing logistic
regression models.
Causal inference and evaluation methods: Introduction to the potential
outcome appraoch (definition of the primitive concepts and of the
assignment mechanism); Randomized experiments; Design and analysis of observational causal studies.
Software: Computer implementation of the studied methods with the software STATA.