## Predicting Cause-Effect Relationships from Incomplete Discrete Observations (1991)

Venue: | SIAM Journal on Discrete Mathematics |

Citations: | 20 - 10 self |

### BibTeX

@ARTICLE{Boros91predictingcause-effect,

author = {E. Boros and P. L. Hammer and J.N. Hooker},

title = {Predicting Cause-Effect Relationships from Incomplete Discrete Observations},

journal = {SIAM Journal on Discrete Mathematics},

year = {1991},

volume = {7},

pages = {531--543}

}

### OpenURL

### Abstract

We address a prediction problem that frequently occurs in practice. We wish to predict the value of a function on the basis of discrete observational dat a that are incomplete in two senses. Only certain arguments of the function ar e observed, and the function value is observed only for certain combinations of values of these arguments. We solve the problem under a monotonicity condition that is natural in many applications, and we discuss applications t o tax auditing, medicine, and real estate valuation. In particular, we display a special class of problems for which the best mono tone prediction can be found in polynomial time. 1 Introduction The problem of establishing cause-effect relationship based on incomplete observations was studied in [4]. In this paper we address the problem of finding a good approximation of an unknown discrete function on the basis of a set of observations, which is incomplete in two senses. We observe the values of only GSIA Working Paper 1991...

### Citations

76 |
Cause-Effect Relationships and Partially Dened Boolean Functions
- Crama, Hammer, et al.
- 1988
(Show Context)
Citation Context ... problems for which the best mono tone prediction can be found in polynomial time. 1 Introduction The problem of establishing cause-effect relationship based on incomplete observations was studied in =-=[4]-=-. In this paper we address the problem of finding a good approximation of an unknown discrete function on the basis of a set of observations, which is incomplete in two senses. We observe the values o... |

40 |
A selection problem of shared fixed costs and network flows
- Rhys
- 1970
(Show Context)
Citation Context ...osure in a given directed graph, and he showed that the maximal closure can be found as the source side of a minimum cut in an associated network. This problem has many interesting applications, e.g. =-=[1, 7, 8, 9, 10, 12, 13, 16]-=-. In this section we introduce a generalization of this problem, which is shown to be equivalent with the problem of V -monotone approximation, and which is reducible in certain cases to the problem o... |

33 |
Maximum closure of a graph and applications to combinatorial problems
- Picard
- 1976
(Show Context)
Citation Context ... statement of the problem and a small example. In Section 3 we show that finding the best monotone approximation is equivalent to a generalization of the "maximal closure" problem on a direc=-=ted graph [11]-=-. When there are two ordered outcomes, the generalization coincides with the maximal closure problem. In Section 4 we show that the generalized maximal closure problem can be reduced to the maximal cl... |

29 |
Assessing credit card applications using machine learning
- Carter, Catlett
- 1987
(Show Context)
Citation Context ...values of f(x) for unobserved x are in general restricted. In [2] we consider some other possible restrictions on f (when it is a boolean function). More examples for similar problems can be found in =-=[3, 14, 15]-=-. In this paper we show that a network flow model can be used to determine the best approximation f when the partial ordering of the outcomes is an interval order. An interval order is one in which wh... |

26 |
Learning categorical decision criteria in biomedical domains
- Spackman
- 1988
(Show Context)
Citation Context ...values of f(x) for unobserved x are in general restricted. In [2] we consider some other possible restrictions on f (when it is a boolean function). More examples for similar problems can be found in =-=[3, 14, 15]-=-. In this paper we show that a network flow model can be used to determine the best approximation f when the partial ordering of the outcomes is an interval order. An interval order is one in which wh... |

24 |
On a selection problem
- Balinski
- 1970
(Show Context)
Citation Context ...osure in a given directed graph, and he showed that the maximal closure can be found as the source side of a minimum cut in an associated network. This problem has many interesting applications, e.g. =-=[1, 7, 8, 9, 10, 12, 13, 16]-=-. In this section we introduce a generalization of this problem, which is shown to be equivalent with the problem of V -monotone approximation, and which is reducible in certain cases to the problem o... |

17 |
Capital Budgeting of Interrelated Projects, Survey and Synthesis
- Weingartner
(Show Context)
Citation Context ...osure in a given directed graph, and he showed that the maximal closure can be found as the source side of a minimum cut in an associated network. This problem has many interesting applications, e.g. =-=[1, 7, 8, 9, 10, 12, 13, 16]-=-. In this section we introduce a generalization of this problem, which is shown to be equivalent with the problem of V -monotone approximation, and which is reducible in certain cases to the problem o... |

12 |
Selected applications of minimum cuts in networks
- Picard, Queyranne
- 1982
(Show Context)
Citation Context ... The closure of a set C of nodes (x's) is the set of all x's greater than or equal to the x's in C. For instance, the 5 Figure 1: Illustration of a V-chain. closure o f f3g is f3; 4; 5g. J.-C. Picard =-=[12]-=- introduced the problem of finding a maximal closure in a given directed graph, and he showed that the maximal closure can be found as the source side of a minimum cut in an associated network. This p... |

11 |
Inductive learning for risk classification
- Shaw, Gentry
- 1990
(Show Context)
Citation Context ...values of f(x) for unobserved x are in general restricted. In [2] we consider some other possible restrictions on f (when it is a boolean function). More examples for similar problems can be found in =-=[3, 14, 15]-=-. In this paper we show that a network flow model can be used to determine the best approximation f when the partial ordering of the outcomes is an interval order. An interval order is one in which wh... |

9 |
Optimizing field repair kits based on job completion rate
- Mamer, Smith
- 1982
(Show Context)
Citation Context |

2 |
Boolean Regression," manuscript
- Boros, Hammer, et al.
- 1991
(Show Context)
Citation Context ...r a given observation, a penalty is incurred, where the size of the penalty depends on the correct and predicted outcomes. For us, the best approximation f is one that minimizes the total penalty. In =-=[2] we invest-=-igate the issue of defining "best approximation" from a statistical point of view and distinguish the analysis described here from logit and categorical data analysis. If there is no restric... |

2 | A New Algorithm for Computing the Maximal Closure of a Graph", Management Science 36 - Faaland, Kim, et al. - 1990 |

2 |
Quelques Approches de la Programmation Non Lin'eaire en variables
- Hansen
- 1974
(Show Context)
Citation Context |

2 |
Optimal Open Pit Mine Contours with Optimal Depletion Scheduling: Theory and Methodology
- Hauck, Malone
- 1969
(Show Context)
Citation Context |

2 |
Rhys's Combinatorial Station Selection Problem", London Graduate School of Business Studies, Transport Network Theory Unit, Report LBS-TNT-68
- Murchland
- 1968
(Show Context)
Citation Context |