Properties of the OLS estimator under numerous patterns of heteroskedasticity

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62156489%3A43110%2F16%3A43909258" target="_blank" >RIV/62156489:43110/16:43909258 - isvavai.cz</a>
Result on the web
<a href="http://dx.doi.org/10.1063/1.4951810" target="_blank" >http://dx.doi.org/10.1063/1.4951810</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1063/1.4951810" target="_blank" >10.1063/1.4951810</a>

Result language
angličtina
Original language name
Properties of the OLS estimator under numerous patterns of heteroskedasticity
Original language description
Favourable properties of the Ordinary Least Squares estimator are known to depend upon compliance with a set of restrictive conditions, often referred to as classical assumptions. This paper discusses evidence from Monte Carlo simulation schemes aimed at verifying occurrence of bias and minimum variance property of OLS estimator in presence of various patterns of heteroskedasticity. Three independent MC simulation schemes were prepared to investigate consequences of error term variance changing with power of the proportionality factor or depending upon the variance multiplication constant applied to a subset of observations. Several finite sample properties of the OLS estimator were shown to vary according to the simulated form of the skedastic function, but to a much lesser scope with the sample size.
Czech name
—
Czech description
—

Project
—
Continuities
I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Publication year
2016
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Similar results(10)