In: Statistics and Probability
Advertising expenses are a significant component of the cost of goods sold. Listed below is a frequency distribution showing the advertising expenditures for 71 manufacturing companies located in the Southwest. The mean expense is $50.28 million and the standard deviation is $11.46 million. Is it reasonable to conclude the sample data are from a population that follows a normal probability distribution?
Advertising Expense ($Million) Number of Companies
25 up to 35 8
35 up to 45 12
45 up to 55 30
55 up to 65 12
65 up to 75 9
Total 71
A.) State the decision rule. Use the 0.10 significance level. (round 2 decimal places)
H0: The population of advertising expenses follows a normal distribution
H1: The population of advertising expenses does not follow a normal distribution
B.) Compute the value of the chi-square. (round 2 decimal places)
Solution:-
For this kind of problem we can plot Histogram of Observed Frquency.
Here,
mean of expense is $50.28 million and standard deviation is $11.46 million
So,using this information we have to conclude that whether the data is actually coming from the normal population.Hence,we will be computing chi-square goodness of fit statistics and thereby conclude the findings.Null and alternative hypothesis for the above testing will be
H0: The population of advertising expenses follows a normal distribution
H1: The population of advertising expenses does not follow a normal distribution
A) Decision rule at 0.10 level of significance.
We will be rejecting the null hypothesis if,
" Chi-sq-calculated > Chi-sq-tabulated(alpha==0.10,df) "
or else
We can compare p-value of this test statistic with the level of significance(0.10)
If p-value > 0.10 then we reject the null hypothesis otherwise we do not sufficient evidence to reject the null hypothesis.
B) Compute the value of chi square.
Necessary formulas:-
We know that in case of the continuous distribution:-
And as E(X) = 50.28
sd(X) = 11.46
Hence it becomes-
Where Z is also standardized normal variable using mean and standard deviation.
So we will evaluate RHS probability using normal table.
Please find the calculations below.
Here,we can see that chi-calculated < chi-table;hence our decision will be to accept the null hypothesis.
Conclusion:- At 0.10 significance level, -We conclude that the data is coming from the normal population.
i.e "The population of advertising expenses follows a normal distribution" at 0.10 significance level.
Which we can see from below histogram of expected frequencies as well.
Both histograms are showing the normality of the frequencies.