Efficient Estimator - Relative Efficiency

Relative Efficiency

If and are estimators for the parameter, then is said to dominate if:

  1. its mean squared error (MSE) is smaller for at least some value of
  2. the MSE does not exceed that of for any value of θ.

Formally, dominates if

\mathrm{E} \left[ (T_1 - \theta)^2 \right] \leq \mathrm{E} \left[ (T_2-\theta)^2 \right]

holds for all, with strict inequality holding somewhere.

The relative efficiency is defined as

e(T_1,T_2) = \frac {\mathrm{E} \left} {\mathrm{E} \left}

Although is in general a function of, in many cases the dependence drops out; if this is so, being greater than one would indicate that is preferable, whatever the true value of .

Read more about this topic:  Efficient Estimator

Famous quotes containing the words relative and/or efficiency:

    “In a country where misery and want were the foundation of the social structure, famine was periodic, death from starvation common, disease pervasive, thievery normal, and graft and corruption taken for granted, the elimination of these conditions in Communist China is so striking that negative aspects of the new rule fade in relative importance.”
    —Barbara Tuchman (1912–1989)

    “Nothing comes to pass in nature, which can be set down to a flaw therein; for nature is always the same and everywhere one and the same in her efficiency and power of action; that is, nature’s laws and ordinances whereby all things come to pass and change from one form to another, are everywhere and always; so that there should be one and the same method of understanding the nature of all things whatsoever, namely, through nature’s universal laws and rules.”
    —Baruch (Benedict)