1 / 11

An Experimental Test of Agreeing to Disagree

An Experimental Test of Agreeing to Disagree. ICES Brown Bag, Mar.1, 2002 Robin Hanson, G.M.U., presenting William Nelson, SUNY Buffalo. Agreeing to Disagree Theory.

fifi
Download Presentation

An Experimental Test of Agreeing to Disagree

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. An Experimental Test of Agreeing to Disagree ICES Brown Bag, Mar.1, 2002 Robin Hanson, G.M.U., presenting William Nelson, SUNY Buffalo

  2. Agreeing to Disagree Theory • E1[x] = E2[x] if x is r.v., finite space, Bayesian, common prior, common knowledge of exact current values E1[x], E2[x] (Aumann 76) • exact values → sep. point (Sebenius & Geanakoplos 83), monotone stat. (McKelvey & Page 86), who has max value (Hanson 98a) • Bayesian → balanced (Geanakoplos 89), know that they know (Samet 90), Turing machine (Medgiddo 89), Bayesian wannabe (Hanson 98b)

  3. Agreeing to Disagree Theory II • finite → infinite state space (Samet 92) • =,c.k. →≈,common belief (Monderer & Samet 89) • current values → eventually if finite, last values honestly said (Geanakoplos & Polemarchakis 82) • E1t[x] = E1t[E2t+s[x]] if finite space, Bayesian, common prior, common knowledge that 2 hears 1 honestly say if E1t[x] ≥ E1t[E2t+s[x]] (Hanson 02) (Should generalize to infinite space, common belief, Bayesian wannabes)

  4. Comparing Theory to Reality • Humans seem to disagree constantly, seem well aware of this, even those know theory • So we not seek truth? (Cowen & Hanson 01) • But many question if theory ever tested GOAL: a more direct agreeing to disagree test

  5. A gets clue XA A1 = A’s guess of X A told Sign(B2-B1) A2 = A’s guess of X Loss (A1-X)2, (A2-X)2 B gets clue XB B told A1 B1 = B’s guess of X B2 = B’s guess of A2 Loss (B1-X)2, (B2-A2)2 Dice Game X = XA + XB time

  6. A1 = A’s guess of X A told Sign(B2-B1) A2 = A’s guess of X Loss (A1-X)2, (A2-X)2 B told A1 B1 = B’s guess of X B2 = B’s guess of A2 Loss (B1-X)2, (B2-A2)2 Percent Game time E.g.: What % of U.S. say dogs better pets than cats?

  7. Percent Forms

  8. Sample Percent Questions • What percent of people in the U.S. agree with this opinion? “God created humans in basically their present form in the last 10,000 years.” (Gallup,1999) • What percent of people in the U.S. agree with this opinion? “The U.S. government is hiding that it knows of the existence of aliens.” (CNN 1994) • By weight, what percent of cheddar cheese is protein? (U.S. Department of agriculture) • What percent of the population of India is literate? (Nation of India)

  9. Experiment Features • All answers integers in [0,100], either real % or XA + XB, each from 6s dice: [0,10,20,30,40,50] • All by hand, subjects roll dice first, for credibility • Subjects told all after each round, to help learning • Zipper design, to minimize strategic interactions • Lottery payoff, to reduce risk aversion • Double dice, for easy squared-error penalty • Only tell B-sign, to reduce signaling ability

  10. Results • * = 5%, ** = 1%, *** = .2% • Exclude B-sign=0 • Really B-sign*(A1-X) … • In both games, A2 neglects info in B-sign, and B2 accurately anticipates this. • In percent game, B1 neglects info in A1 • No correlation with self-deception measure, question emotion level

  11. Where go from here? • Do we need more data of same form? • Should we see degree of signaling by also paying on estimates never show others? • Should see degree of common belief via auctions pay off on subject test error rate?

More Related