|
Teaching GuideTerm Faculty of Computer Science |
| Mestrado Universitario en Técnicas Estadísticas (Plan 2019) |
 Subjects |
| Nonparametric Methods |
| Contents |
| Identifying Data | 2019/20 | |||||||||||||
| Subject | Nonparametric Methods | Code | 614493111 | |||||||||||
| Study programme |
|
|||||||||||||
| Descriptors | Cycle | Period | Year | Type | Credits | |||||||||
| Official Master's Degree | 1st four-month period |
First | Obligatory | 5 | ||||||||||
|
||||||||||||||
| Topic | Sub-topic |
| Nonparametric distribution estimation |
The empirical distribution. Properties. Moments and quantiles estimation. |
| Classical one-sample nonparametric tests. | Goodness-of-fit tests: Kolmogorov-Smirnov test. Normality analysis: Q-Q plot, Lilliefors test, Shapiro-Wilk test, transformations for normality. One-sample location tests: sign test, Wilcoxon signed-rank test. |
| Two-sample tests. |
Two-sample comparison: Kolmogorv-Smirnov test for two-samples, Mann-Whitney-Wilcoxon test. Extensions for three or more samples: Kruskal-Wallis test, Friedman test. |
| Tests based on contingency tables. | Contingency tables analysis. Chi-squared tests for goodness-of-fit, homogeneity and independence on contingency tables. |
| Smoothing methods: nonparametric density estimation. | The histogram. Kernel density estimation. Assessment of density estimators. Smoothing parameter selectors in kernel density estimation: cross-validation and plug-in approaches. Multivariate kernel density estimation. |
| Nonparametric regression estimation. | Kernel regression. Local polynomial regression. k-nearest neighbor regression. Smoothing parameter selectors in kernel regression estimation: cross-validation and plug-in approaches. Loess algorithm. Spline regression: a brief introduction. |
|
