Saddle Point
Back to the cover

Part I · Foundations

  1. 0 Mathematical Foundations
  2. 1 Observation, Beliefs, and Decision

Part II · Engines

  1. 2 Competing for Allocation: The Blotto Game
  2. 3 Bargaining, Commitment, and Repeated Cooperation
  3. 4 Signals, Disclosure, and Persuasion
  4. 5 Different Models of One Game
  5. 6 Leadership and Rules: Inspection
  6. 7 Decisions in Time
  7. 8 Many Agents
  8. 9 Networks, Dependence, and Resilience
  9. 10 Aggregating Information

Part III · Systems

  1. 11 Architecture of Decision Support
  2. 12 Simulation and Wargaming with AI Agents

Part IV · Limits

  1. 13 Where the Mathematics Stops

Chapter 10 · Aggregating Information

Leonard J. Savage

1917 – 1971

The problem on the desk

How can a person’s probabilities be elicited through choices or payments? Savage examined conditions under which reporting beliefs honestly maximises expected utility, while separating the mathematics from practical limitations.1

A life in moments

  1. 1917

    Detroit

    Born on 20 November.2

  2. 1941

    University of Michigan · Ann Arbor

    Completed his doctorate.3

  3. 1946–1947

    Rockefeller Foundation fellowship; Woods Hole and University of Chicago

    Held a fellowship before joining Chicago’s staff.3

  4. 1954

    University of Chicago · Chicago

    Published The Foundations of Statistics.3

  5. 1960

    University of Michigan · Ann Arbor

    Moved to Michigan.3

  6. 1964

    Yale University · New Haven

    Joined Yale.3

  7. 1971

    New Haven

    Died on 1 November, while the elicitation paper was in press.2

Every line rests on a dated source (numbered below) unless it carries a mark: R for an inference, S for a story as told.

In their own words

A probability is a price, in a manner of speaking.

Elicitation of Personal Probabilities and Expectations, 19715

A story

A joint note became a solo paper

Savage describes starting with de Finetti in spring 1960. As the project expanded and de Finetti published related material, they agreed that Savage would publish this version alone.4

Around Savage

  • built on the work Glenn W. Brier Savage explicitly cites Brier’s scoring paper as an earlier special case.4
  • collaboration Bruno de Finetti The elicitation project began as a collaboration with de Finetti.6

Recorded as advisor Sumner Byron Myers Recorded as students Morris Herman DeGroot, Donald Arthur Berry, William Swain Cleveland Mathematics Genealogy Project record

The genealogy lists only what the database records; a missing name is not a missing teacher.

Two desks

Savage and Brier Evaluating reported probabilities. The same problem from two desks →

The card in the book

Statistician whose The Foundations of Statistics built the modern theory of subjective probability and personal decision on axioms of coherent preference. He taught at the University of Chicago (1946–60), the University of Michigan (1960–64) and Yale (1964–71). His paper on elicitation, published posthumously in December 1971, characterised the scoring rules under which announcing one's true probability is optimal — the propriety theory this chapter instantiates.

Main works The Foundations of Statistics (1954); “Elicitation of personal probabilities and expectations” (Journal of the American Statistical Association 66, 1971).

To read further

Sources (6)
  1. Elicitation of Personal Probabilities and Expectations, Leonard J. Savage, 1971; pp. 783–785
  2. Savage, Leonard Jimmie, Barry M. Hill; Dictionary of Scientific Biography; opening and career
  3. Leonard Jimmie Savage Papers, MS 695, Andrew M. Patterson; Yale University Library, 1974; p. 4
  4. Elicitation of Personal Probabilities and Expectations, Leonard J. Savage, 1971; p. 783, §1.1
  5. Elicitation of Personal Probabilities and Expectations, Leonard J. Savage, 1971; p. 783, §1.2
  6. Elicitation of Personal Probabilities and Expectations, Leonard J. Savage, 1971; p. 783

The facts on this page were gathered for the site from the sources listed, each checked against its locator; the book's card is reproduced unchanged.