<section name="raw">
    <SEQUENTIAL>
      <record key="001" att1="001" value="LIB900218605" att2="LIB900218605">001   LIB900218605</record>
      <field key="037" subkey="x">englisch</field>
      <field key="050" subkey="x">Forschungsbericht</field>
      <field key="076" subkey="">Ökonomie</field>
      <field key="079" subkey="y">http://www.ihs.ac.at/publications/ihsfo/fo98.pdf</field>
      <field key="079" subkey="z">bowden, roger j., some approaches to the estimation and discounting of interpolated observations (pdf)</field>
      <field key="100" subkey="">bowden, roger j.</field>
      <field key="103" subkey="">university of auckland</field>
      <field key="331" subkey="">some approaches to the estimation and discounting of interpolated observations</field>
      <field key="403" subkey="">1. ed.</field>
      <field key="410" subkey="">wien</field>
      <field key="412" subkey="">institut fuer hoehere studien</field>
      <field key="425" subkey="">1976, march</field>
      <field key="433" subkey="">31 pp.</field>
      <field key="451" subkey="">institut fuer hoehere studien; forschungsberichte; 98</field>
      <field key="544" subkey="">IHSFO 98</field>
      <field key="753" subkey="">abstract: the interpolation of economic time series has a long and consistently unfashionable history as an object of study.</field>
      <field key="nev" subkey="e">rtheless economists have from time to time been forced by the inadequacy of their data or the ambition of their constructs to</field>
      <field key="con" subkey="s">ider ways of filling in the gaps where observations are missing. the problem known as "missing observations" has been fairly</field>
      <field key="int" subkey="e">nsively researched in the statistical literature (for a review, see a.a. afifi and e.m. elashoff (1966)). the tenor of this</field>
      <field key="wor" subkey="k">has been to design parameter estimators which compensate, as it were, for the missing data, rather than stressing the</field>
      <field key="est" subkey="i">mation of the missing observations themselves. the economist, however, is often interested in testing a numberof alternative</field>
      <field key="hyp" subkey="o">theses, perhaps of differing functional form. to design a different estimation procedure for each particular hypothesis under</field>
      <field key="tes" subkey="t">would represent a substantial and possibly unjustified use of research time. thus it is useful to be able to estimate in a</field>
      <field key="rel" subkey="a">tively simple way the missing observations themselves, and once having achieved the complete series, to know what</field>
      <field key="mod" subkey="i">fications must be made to standard tests in screening a number of alternative hypotheses. there are thustwo related tasks to</field>
      <field key="whi" subkey="c">h the present paper addresses itself. the first is to find a relatively simple and general method of estimation of the</field>
      <field key="obs" subkey="e">rvations to be interpolated. the use of a related series for which observations are readily availableon the required basis</field>
      <field key="was" subkey="">investigated by m. friedman (1962). g.c. chow and an-loh lin (1971) showed how to formalize and generalize this work in the</field>
      <field key="con" subkey="t">ext of least-squares estimators. to complement this work we need now to pay some attention to methods based on likelihood</field>
      <field key="pri" subkey="n">ciples. the method of maximum-likelihood was applied by m. drettakis (1973) in a missing-observations study, to generate</field>
      <field key="est" subkey="i">mates of observations missing at the start of a series for a simultaneous equation model. in the present paper we study the</field>
      <field key="mor" subkey="e">usual situation where observations are missing in a regular way throughout the series. although we investigate briefly a</field>
      <field key="met" subkey="h">od based on true maximum likelihood principles, we reject at in favour of a "quasi-likelihood" procedure which perhaps has</field>
      <field key="mor" subkey="e">in common with bayesian techniques. given that one has obtained point estimates with associated variances or covariances,</field>
      <field key="the" subkey="r">e remains the problem of how the now-complete data series is to be deployed. under the</field>
    </SEQUENTIAL>
  </section>