In: Statistics and Probability
The number of faults over a period of time was collected for a sample of 100 data transmission lines. We want to test if the data come from a Poisson distribution.
number of faults | 0 | 1 | 2 | 3 | 4 | 5 | >5 |
number of lines | 38 | 30 | 16 | 9 | 5 | 2 | 0 |
(a) Assuming the number of faults for a data-transmission line, Yi , i = 1, . . . , 100, follows a Poisson distribution with parameter λ, find the maximum likelihood estimate of λ for these data.
(b) Test the hypothesis that the number of faults for a data-transmission line follows a Poisson distribution using the chi-squared test. Do we reject the null hypothesis at the 5% significance level? Hint: You should combine the observed data into several groups such that expected frequencies are greater or equal to 5 for each group.