Class B insulation. Use the 220°C data on (log) time to failure of insulation specimens in Table 3.1 of Chapter 7.
(a) Make a lognormal plot of the data, and graphically estimate the lognormal parameters. Using the estimate of u. describe whether the failure rate increases or decreases with age.
(b) (b) Calculate the ML estimates of the parameters and the lognormal median. The referenced tables for singly censored samples may help reduce the effort. as would a computer program.
(c) Calculate the sample local Fisher information matrix for the parameter estimates.
(d) Calculate the local estimate of the asymptotic covariance matrix of the parameter estimates.
(e) Calculate two-sided approximate 95% confidence limits for the parameters and lognormal median.
(f) Calculate the ML estimate for the 10th percentile of the lognormal distribution.
(g) Calculate the local estimate of the asymptotic variance of the log of the ML estimator (f) of the 10th percentile.
(h) Calculate two-sided approximate 95% confidence limits for the 10th (lognormal) percentile
(i) Using (b) and (f), plot the ML fitted distribution on the lognormal plot from (a). Also, plot the confidence limits from (e) and (h)
(j) Calculate limits for the parameters and lognormal median using exact tables referenced in Section 2.
(k) Repeat (c) through (h) using the first-derivative estimate of Section 6.3 for the Fisher information matrix.
(l) 
Table
