| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 339.17 KB | Adobe PDF |
Advisor(s)
Abstract(s)
We consider selecting a regression model, using a variant of the generalto-
specific algorithm in PcGets, when there are more variables than observations. We
look at the special case where the variables are single impulse dummies, one defined for
each observation. We show that this setting is unproblematic if tackled appropriately,
and obtain the asymptotic distribution of the mean and variance in a location-scale
model, under the null that no impulses matter. Monte Carlo simulations confirm the
null distributions and suggest extensions to highly non-normal cases
Description
Keywords
Indicators Regression saturation Subset selection Model selection
Pedagogical Context
Citation
SANTOS, Carlos; HENDRY, David F.; JOHANSEN, Soren - Automatic selection of indicators in a fully saturated regression. Computational Statistics. ISSN: 1613-9658. Vol.23 (2008) p. 317–335
Publisher
Springer