Page 499 - Probability and Statistical Inference
P. 499

476    9. Confidence Interval Estimation

                                    9.4.3 (Exercise 9.4.2 Continued) Suppose that X , ..., X  are iid random
                                                                             i1
                                                                                   in
                                                      2
                                 samples from the N(µ , σ ) population, i = 1, ..., 5. The observations X , ...,
                                                    i
                                                                                             i1
                                              th
                                 X  refer to the i  treatment, i = 1, ..., 5. Let us assume that the treatments are
                                  in
                                 independent too and that all the parameters are unknown. With fixed α ∈ (0,
                                 1), derive the (1 – α) joint confidence intervals for estimating the parameters
                                 θ  = µ  – µ , θ  = µ  – µ  and θ  = µ  + µ  – 2µ .
                                  1   1    2  2   2   3     3    3   4    5
                                    9.4.4 Use the equation (9.4.20) to propose the simultaneous (1 – α) two-
                                 sided confidence intervals for the variance ratios.
                                    9.4.5 Let X , ..., X  be iid having the common pdf f(x; µ , σ ), n ≥ 2,
                                                                                       i
                                                    in
                                              i1
                                                                                          i
                                 where f(x; µ, σ) = σ  exp{–(x – µ)/σ}I(x > µ), (µ , σ ) ∈ ℜ × ℜ , i = 0, 1,
                                                   –1
                                                                                         +
                                                                               i
                                                                            i
                                 ..., p. The observations X , ..., X  refer to a control whereas X , ..., X in
                                                                                          i1
                                                               0n
                                                       01
                                 refer to the i  treatment, i = 1, ..., p. Suppose that all the observations are
                                            th
                                 independent and all the parameters are unknown. Start with an appropriate p–
                                 dimensional pivot depending on the sufficient statistics and then, with fixed α
                                 ∈ (0, 1), derive the simultaneous (1 – α) two-sided confidence intervals for
                                 the ratios σ /σ , i = 1, ..., p. {Hint: Proceed along the lines of the Examples
                                           i
                                             0
                                 9.4.6–9.4.7.}
                                    9.4.6 (Exercise 9.3.10 Continued) Suppose that the random variables X ,
                                                                                                i1
                                 ..., X  are iid having the common pdf f(x; θ ) where we denote the exponen-
                                                                      i
                                     in
                                 tial pdf f(x; a) = a  exp{–x/a}I(x > 0), θ  > 0, i = 0, 1, ..., p. The observations
                                                –1
                                                                   i
                                                                                      th
                                 X , ..., X  refer to a control whereas X , ..., X  refer to the i  treatment, i
                                  01
                                                                   i1
                                                                          in
                                         0n
                                 = 1, ..., p. Suppose that all the observations are independent and all the param-
                                 eters are unknown. Start with an appropriate p–dimensional pivot depending
                                 on the sufficient statistics and then, with fixed α ∈ (0, 1), derive the simulta-
                                 neous (1 – α) two-sided confidence intervals for the ratios θ /θ , i = 1, ..., p.
                                                                                    i
                                                                                      0
                                 {Hint: Proceed along the lines of the Examples 9.4.6–9.4.7.}
   494   495   496   497   498   499   500   501   502   503   504