We deal here with a multinomial experiment. Note that your data set is not completed as you have only five categories (number of errors) and the observed frequencies for the sample of n = 500. You need to add to this data the historical percentage the manual invoicing system produced. a. Develop the chart like below and transfer the data from the data set provided as well as the information from the case. As we don't use percentages in computing enter the corresponding probabilities in decimals.

b. Now use the probabilities from the chart to set the null and alternative hypothesis to test whether the error percentages for the computerized system differ from those for the manual system (Goodness-of-Fit Test for Multinomial Probabilities).

Invoicing System Data
Errors	Computerized Frequencies
0	479
1	10
2	8
3	2
>3	1

ComputerizedInvoicingSystemDataAnalysis

	Computerized	Manual
Errors	ObservedFrequencies, f	PreviousProbabilities,p
0
...
Total