Out computer scientist now tests her algorithm on a sample of 400 scanned x-ray images.
The results of the sample suggest that the general mean processing time is 3505 with a variance of 100. Should she reject her earlier belief that the mean processing time is 3500 milliseconds?
Can reject at 0.1% significance
Can only reject at 1% significance
Can only reject at 5% significance
Cannot reject earlier belief
Not enough information to answer this question
What are the units of the variance?
In a separate test, the memory usage of our computer scientist’s image processing algorithm is tested against the memory usage of other algorithms, and against memory usage when different types of image content are considered. The results on an ANOVA test are shown below. The factor Algorithm refers to the choice of algorithm used, and the factor Content refers to different types of image content.
√milliseconds
milliseconds
milliseconds2
milliseconds3
none of the above
How many different types of image processing algorithms were tested in this ANOVA?
2
3
4
5
Not enough information to answer this question
From the ANOVA results, which factors would you say were significantly related to memory usage at 5% significance or less?
The choice of algorithm
The choice of image content
The interaction of algorithm and content
More than one of the above
None of the above
From the ANOVA results, how much of the total variation is due to differences in the algorithm (to the nearest percent)?
A sport coach is monitoring the performance (recorded as points) of an athlete every day at practice, and has also kept record of how much sleep (recorded as hours) he got the night before practice, and how much caffeine (recorded in milliliters) he consumed before practice each day. Each row in the coach’s dataset represents one day’s entry of performance, sleep, and caffeine intake. The coach has conducted a regression where the dependent variable was performance, and the independent variables sleep and caffeine. An interaction between sleep and caffeine was also included. All variables were fully standardized before running the regession.
22%
33%
55%
74%
Not enough information to answer this question
We are only interested in factors that are significant at 5% significance or less. Which independent terms should we consider to be significant?
Only sleep
Only caffeine
Only Sleep*Caffeine
Sleep and also Sleep*Caffeine
Caffeine and also Sleep*Caffeine
What are the units of the estimate of the sleep coefficient reported in the results above?
no units
points/hour
hours/point
hours2
None of the above
How would you interpret the regression coefficient of the interaction term?
Interaction is not significant-we can ignore it.
The relationship between sleep and performance decreases as caffeine intake increases
The relationship between caffeine and performance decreases as sleep intake increases
Bothe (B) and (C) are valid
The interaction is significant but neither (B) nor (C) are valid
How many days worth of data would you say the coach has recorded?
50 days
60 days
70 days
80 days
Not enough information to answer this question
If the athlete in question kept his caffeine intake to his average, but increased his amount of sleep by one standard deviation, by how would you predict his performance would change?
Performance would increase by 0.23 standard deviations
Performance would increase by 0.25 standard deviations
Performance would increase by 0.43 standard deviations
Performance would increase by 1 standard deviation
Not enough information to answer this question
From our regression, how much of the variance of performance should we report is predicted by sleep and caffeine?
22.87%
25.8%
53.66%
87.82%
Not enough information to answer this question
可觀看題目詳解,並提供模擬測驗!(免費會員無法觀看研究所試題解答)