In: Math
One measure of the value of a stock is its price to earnings ratio (or P/E ratio). It is the ratio of the price of a stock per share to the earnings per share and can be thought of as the price an investor is willing to pay for $1 of earnings in a company. A stock analyst wants to know whether the P/E ratios for three industry categories differ significantly. The following data represent simple random samples of companies from three categories: (1) financial, (2) food, and (3) leisure goods.
Financial |
Food |
Leisure Goods |
8.83 |
19.75 |
14.1 |
12.75 |
17.87 |
10.12 |
13.48 |
15.18 |
15.57 |
14.42 |
22.84 |
13.48 |
10.06 |
15.6 |
11.27 |
H0: ______________________________ versus H1: _________________________________________.
H0: __________________ versus H1: ____________________.
q0=
Decision:
H0: __________________ versus H1: ____________________.
q0=
Decision:
H0: __________________ versus H1: ____________________.
q0=
Decision:
• Use lines to indicate which population means are not significantly different.
From above probability plot we observed that the observations come from normal distribution.
Test for Equal Variances: P/E ratio versus categories
95% Bonferroni confidence intervals for standard deviations
categories N
Lower StDev Upper
Financial 5
1.27831 2.36535 9.1040
Food 5 1.70707 3.15871
12.1576
Leisure Goods 5 1.18681 2.19604 8.4523
Bartlett's Test (Normal Distribution)
Test statistic = 0.56, p-value = 0.757
Levene's Test (Any Continuous Distribution)
Test statistic = 0.28, p-value = 0.760
Since p-value>0.05 so we can assume population variances for three categories are same.
Now we construct ANOVA table:
ANOVA | ||||||
Source of Variation | SS | df | MS | F | P-value | F crit |
Between Groups | 116.185333 | 2 | 58.09267 | 8.545171 | 0.004927135 | 3.885293835 |
Within Groups | 81.57964 | 12 | 6.798303 | |||
Total | 197.764973 | 14 |
MST=mean square due to treatments/groups=58.0927
MSE=mean square due to error=6.7983
Test statistic=F-ratio=MST/MSE=8.5452
Critical value=3.8853
Since F-ratio>Critical value or p-value=0.0049<0.05 hence we reject null hypothesis at 5% level of significance and conclude that there is sufficient evidence that at least one mean P/E ratio is different from other means P/E ratio.