1: \begin{abstract}

2: This paper focuses on generalizing quantiles from the ordering point of

3: view. We propose the concept of \textit{partial quantiles}, which are

4: based on a given partial order. We establish that partial quantiles

5: are equivariant under order-preserving transformations of the data,

6: robust to outliers, characterize the probability distribution if the

7: partial order is sufficiently rich, generalize the concept of efficient

8: frontier, and can measure dispersion from the partial order

9: perspective.

10: 

11: We also study several statistical aspects of partial quantiles. We

12: provide estimators, associated rates of convergence, and asymptotic

13: distributions that hold uniformly over a continuum of quantile indices.

14: Furthermore, we provide procedures that can restore monotonicity

15: properties that might have been disturbed by estimation error,

16: establish computational complexity bounds, and point out a

17: concentration of measure phenomenon (the latter under independence and

18: the componentwise natural order).

19: 

20: Finally, we illustrate the concepts by discussing several theoretical

21: examples and simulations. Empirical applications to compare intake

22: nutrients within diets, to evaluate the performance of investment

23: funds, and to study the impact of policies on tobacco awareness are

24: also presented to illustrate the concepts and their use.

25: \end{abstract}