Estimation of Parameters From Incomplete Data

Author(s):

Lord, Frederic M.

Publication Year:

1954

Report Number:

RB-54-18

Source:

ETS Research Bulletin

Document Type:

Report

Page Count:

11

Subject/Key Words:

Estimation (Mathematics), Incomplete Data, Maximum Likelihood Statistics, Statistical Analysis, Test Theory

Abstract

Maximum likelihood estimators are found for the parameters of a normal trivariate population in the case where observations on one variable are missing from part of the data and observations on another variable are missing from the remainder of the data. Results are presented showing the amount of additional information obtained by these maximum likelihood methods, which make optimum use of the fragmentary data. The formulas derived may also be applied fragmentary bivariate data.

Read More

• Request Copy (specify title and report number, if any)
• http://dx.doi.org/10.1002/j.2333-8504.1954.tb00245.x