A Pause for Reflection — Question 4

Section 44-9 · Physics pages 7458–7459

Question 4 Competing angular models can be compared with normalized probabilities Two ideal models for x=cos(theta) on [-1,1] have densities fA = 1/2 and fB = 3(1 + x2)/8. Ignore detector acceptance. cos θ A B Schematic; labels, not drawing lengths, specify the data. (a) Integrate each density over the central interval |x|<1/2. (b) For 1600 events, find predicted central counts and the difference in units of the binomial standard deviation under A.
Show solutionHide solution
Question 4 - Solution Competing angular models can be compared with normalized probabilities cos θ A B (a) Method and calculation PA = ∫.5 −.5 1 2 dx = .5 . PB = 3 8[x + x3/3].5 −.5 = 13/32 = .40625 . (b) Interpretation and further calculation NA = 800, NB = 650, ∆N = 150 . σA = √ 1600(.5)(.5) = 20, ∆N/σA = 7.5 . Physical check This is an expected separation for specified ideal models, not an observed discovery significance with systematic effects.

Question 4

Competing angular models can be compared with normalized probabilities

Two ideal models for x=cos(theta) on [-1,1] have densities fA=1/2f_A=1/2 and fB=3(1+x2)/8f_B=3(1+x^2)/8. Ignore detector acceptance.

Diagram shown in the original SVG worksheet.

  1. Integrate each density over the central interval |x|<1/2.

  2. For 1600 events, find predicted central counts and the difference in units of the binomial standard deviation under A.

Question 4 - Solution

Competing angular models can be compared with normalized probabilities

Diagram shown in the original SVG worksheet.

(a) Method and calculation

PA=∫−.5.512dx=.5.\boxed{P_A=\int_{-.5}^{.5}\frac12\,dx=.5}.PB=38[x+x3/3]−.5.5=13/32=.40625.\boxed{P_B=\frac38[x+x^3/3]_{-.5}^{.5}=13/32=.40625}.

(b) Interpretation and further calculation

NA=800,NB=650,ΔN=150.\boxed{N_A=800,\quad N_B=650,\quad\Delta N=150}.σA=1600(.5)(.5)=20,ΔN/σA=7.5.\boxed{\sigma_A=\sqrt{1600(.5)(.5)}=20,\quad\Delta N/\sigma_A=7.5}.

Physical check

This is an expected separation for specified ideal models, not an observed discovery significance with systematic effects.