This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Statistics I: Professor Jun Nie Homework Assignment 2 Section 22 2, 4,8,12,16,22,26 2. No. Though she has the frequency distribution of the sample, it is impossible to recreate the original sample data. With more information, such as standard deviation, a list of outliers, etc, she could probably construct similar data, but not exactly. 4. The advantage of relative frequency distribution is more useful when comparing multiple data sets because the total frequency per dataset may not be the same. Relative frequency distribution allows for the measure of frequency to be standardized to percent, allowing for more representative comparison between two datasets. For example, set A has total frequency 100. Set B has total frequency 200. If one class in A has a frequency of 50 and the respective class in B has frequency 100, the class in set B appears to be much larger than in A. When compared with relative frequency distribution, both classes have a relative frequency of 50%. 8. class width: Class boundaries: 0.000 = (9.99  0.00) / 10 = 1 (.99 + 1.00) / 2 = .995 (1.99 + 2.00) / 2 = 1.995 (2.99 + 3.00) / 2 = 2.995 (3.99 + 4.00) / 2 = 3.995 (4.99 + 5.00) / 2 = 4.995 (5.99 + 6.00) / 2 = 5.995 (6.99 + 7.00) / 2 = 6.995 (7.99 + 8.00) / 2 = 7.995 (8.99 + 9.00) / 2 = 8.995 9.995 Class midpoints: (0.99 + 0.00) / 2 = .5 (1.99 + 0.99) / 2 = 1.5 (2.99 + 1.99) / 2 = 2.5 (3.99 + 2.99) / 2 = 3.5 (4.99 + 3.99) / 2 = 4.5 (5.99 + 4.99) / 2 = 5.5 (6.99 + 5.99) / 2 = 6.5 (7.99 + 6.99) / 2 = 7.5 (8.99 + 7.99) / 2 = 8.5 (9.99 + 8.99) / 2 = 9.5 12. The gap between the lowest weights and the highest weights indicates that there might be 12....
View
Full
Document
This note was uploaded on 08/27/2008 for the course ECON V31.0018 taught by Professor Hall during the Spring '08 term at NYU.
 Spring '08
 Hall

Click to edit the document details