Presentation is loading. Please wait.

Presentation is loading. Please wait.

1 LIMITS Why limits? Methods for upper limits Desirable properties Dealing with systematics Feldman-Cousins Recommendations.

Similar presentations


Presentation on theme: "1 LIMITS Why limits? Methods for upper limits Desirable properties Dealing with systematics Feldman-Cousins Recommendations."— Presentation transcript:

1 1 LIMITS Why limits? Methods for upper limits Desirable properties Dealing with systematics Feldman-Cousins Recommendations

2 2 WHY LIMITS? Michelson-Morley experiment  death of aether HEP experiments CERN CLW (Jan 2000) FNAL CLW (March 2000) Heinrich, PHYSTAT-LHC, “Review of Banff Challenge”

3 3 SIMPLE PROBLEM? Gaussian ~ exp{-0.5*(x-μ) 2 /σ 2 } No restriction on μ, σ known exactly μ  x 0 + k σ BUT Poisson {μ = sε + b} s ≥ 0 ε and b with uncertainties Not like : 2 + 3 = ? N.B. Actual limit from experiment = Expected (median) limit

4 4 Bayes (needs priors e.g. const, 1/μ, 1/ √ μ, μ, …..) Frequentist (needs ordering rule, possible empty intervals, F-C) Likelihood (DON’T integrate your L) χ 2 ( σ 2 =μ) χ 2 ( σ 2 = n) Recommendation 7 from CERN CLW: “Show your L” 1) Not always practical 2) Not sufficient for frequentist methods Methods (no systematics)

5 5 Bayesian posterior  intervals Upper limit Lower limit Central interval Shortest

6 6 90% C.L. Upper Limits x  x0x0

7 7 Ilya Narsky, FNAL CLW 2000

8 8 DESIRABLE PROPERTIES Coverage Interval length Behaviour when n < b Limit increases as σ b increases

9 9 Δln L = -1/2 rule If L (μ) is Gaussian, following definitions of σ are equivalent: 1) RMS of L ( µ ) 2) 1/√(-d 2 L /d µ 2 ) 3) ln( L (μ±σ) = ln( L (μ 0 )) -1/2 If L (μ) is non-Gaussian, these are no longer the same “ Procedure 3) above still gives interval that contains the true value of parameter μ with 68% probability ” Heinrich: CDF note 6438 (see CDF Statistics Committee Web-page) Barlow: Phystat05

10 10 COVERAGE How often does quoted range for parameter include param’s true value? N.B. Coverage is a property of METHOD, not of a particular exptl result Coverage can vary with Study coverage of different methods of Poisson parameter, from observation of number of events n Hope for: Nominal value 100%

11 11 COVERAGE If true for all : “correct coverage” P< for some “undercoverage” (this is serious !) P> for some “overcoverage” Conservative Loss of rejection power

12 12 Coverage : L approach (Not frequentist) P(n, μ) = e -μ μ n /n! (Joel Heinrich CDF note 6438) -2 lnλ< 1 λ = P(n,μ)/P(n,μ best ) UNDERCOVERS

13 13 Frequentist central intervals, NEVER undercovers (Conservative at both ends)

14 14 Feldman-Cousins Unified intervals Frequentist, so NEVER undercovers

15 15 Probability ordering Frequentist, so NEVER undercovers

16 16  = (n- µ) 2 /µ Δ = 0.1 24.8% coverage? NOT frequentist : Coverage = 0%  100%

17 17 COVERAGE N.B. Coverage alone is not sufficient e.g. Clifford (CERN CLW, 2000) “Friend thinks of number Procedure for providing interval that includes number 90% of time.”

18 18 COVERAGE N.B. Coverage alone is not sufficient e.g. Clifford (CERN CLW, 2000) Friend thinks of number Procedure for providing interval that includes number 90% of time. 90%: Interval = -  to +  10%: number = 102.84590135…..

19 19 INTERVAL LENGTH Empty  Unhappy physicists Very short  False impression of sensitivity Too long  loss of power (2-sided intervals are more complicated because ‘shorter’ is not metric- independent: e.g. 0  4 or 4  9)

20 20 90% Classical interval for Gaussian σ = 1 μ ≥ 0 e.g. m 2 (ν e )

21 21 Behaviour when n < b Frequentist: Empty for n < < b Frequentist: Decreases as n decreases below b Bayes: For n = 0, limit independent of b Sen and Woodroofe: Limit increases as data decreases below expectation

22 22 FELDMAN - COUSINS Wants to avoid empty classical intervals  Uses “ L -ratio ordering principle” to resolve ambiguity about “which 90% region?”  [Neyman + Pearson say L -ratio is best for hypothesis testing] Unified  No ‘Flip-Flop’ problem

23 23 X obs = -2 now gives upper limit

24 24

25 25

26 26 Black lines Classical 90% central interval Red dashed: Classical 90% upper limit Flip-flop

27 27

28 28 Poisson confidence intervals. Background = 3 Standard Frequentist Feldman - Cousins

29 29

30 30

31 31

32 32

33 33

34 34

35 35

36 36

37 37

38 38

39 39

40 40

41 41

42 42 Recommendations? CDF note 7739 (May 2005) Decide method in advance No valid method is ruled out Bayes is simplest for incorporating nuisance params Check robustness Quote coverage Quote sensitivity Use same method as other similar expts Explain method used

43 43

44 44

45 45 Caltech Workshop, Feb 11th

46 46

47 47

48 48

49 49

50 50

51 51

52 52

53 53

54 54

55 55 Tomorrow is last day of this visit Contact me at: l.lyons@physics.ox.ac.uk

56 56 Peasant and Dog 1)Dog d has 50% probability of being 100 m. of Peasant p 2)Peasant p has 50% probability of being within 100m of Dog d dp x River x =0River x =1 km

57 57 Given that: a) Dog d has 50% probability of being 100 m. of Peasant, is it true that: b ) Peasant p has 50% probability of being within 100m of Dog d ? Additional information Rivers at zero & 1 km. Peasant cannot cross them. Dog can swim across river - Statement a) still true If Dog at –101 m, Peasant cannot be within 100m of Dog Statement b) untrue


Download ppt "1 LIMITS Why limits? Methods for upper limits Desirable properties Dealing with systematics Feldman-Cousins Recommendations."

Similar presentations


Ads by Google