Business Analytics - MIDTERM
CME Group Foundation Business Analytics Lab
Jordan Riebe
7/24
Question 1: Read in the gambling dataset check the first couple of rows and describe the data types. Identify incorrect data types, if any. ( 5 Points )
The datatype for sex seems to be incorrect because it should be nominal listing as “male, female, and other” rather than just a list of “1’s”. The status income is listed as either interval or ratio depending on if there is a true zero. By looking at the numbers it looks like status might mean age, which would make it a ratio datatype. Verbal is an ordinal datatype because they are ranked but there isn’t consistent, meaningful difference between numbers. The datatype for gamble is ratio because there are meaningful, consistent differences between the numbers and there is a true zero. Income is also a ratio datatype because you can earn zero income and there is a meaningful consistent relationship between dollar amounts.
mydata = read.csv(file="data/gambling.csv")
mydataQuestion 2: Describe the data using full sentences and using descriptive statistics. ( 5 Points )
The average score for how verbal a participant is, is 6.65. The average score for gambling is 19.30. The miniumum score for “gamble” is 0, meaning some players are not gambling. The minimum score for “verbal” is 1, which means at least one player is very quiet.The range of status is 18 to 75 meaning there is a 57 number differnce between highest and lowest status.
meanverbal [1] 6.659574verbal = mydata$verbal
verbal [1] 8 8 6 4 8 6 7 5 6 7 6 6 4 6 6 8
[17] 8 5 8 9 8 9 5 4 7 7 4 6 7 8 2 7
[33] 7 10 1 8 7 6 6 6 9 9 8 9 6 7 9gamble = mydata$gamble
meangamble = mean(gamble)
meangamble[1] 19.30106mingamble = min(gamble)
mingamble[1] 0minverbal = min(verbal)
minverbal[1] 1status = mydata$status
status [1] 51 28 37 28 65 61 28 27 43 18 18 43 30 28 38 38
[17] 28 18 43 51 62 47 43 27 71 38 51 38 51 62 18 30
[33] 38 71 28 61 71 28 51 65 48 61 75 66 62 71 71rangestatus = range(status)
rangestatus[1] 18 75Question 3: Estimate the upper and lower threshold for the verbal score ( 5 Points )
The upper threshold would be 12.22925 The lower threshold would be 1.0899
HINT: A common way to estimate the upper and lower threshold is to take the mean (+ or -) 3 * standard deviation.
verbal = mydata$verbal
verbal [1] 8 8 6 4 8 6 7 5 6 7 6 6 4 6 6 8
[17] 8 5 8 9 8 9 5 4 7 7 4 6 7 8 2 7
[33] 7 10 1 8 7 6 6 6 9 9 8 9 6 7 9meanverbal = mean(verbal)
meanverbal[1] 6.659574sdverbal = sd(verbal)
sdverbal[1] 1.856558upperthreshold = mean(verbal) + 3 * sd(verbal)
upperthreshold[1] 12.22925lowerthreshold = mean(verbal) - 3* sd(verbal)
lowerthreshold[1] 1.0899Question 4: Calculate the z-score for income where x=13. Based on the income value x=13 pounds per week, how would you rate the income: low income, average income, high income. Why? ( 5 Points )
I would rate this as a high income becuase the z score came out to be 2.3534 which means that an income of 13 pounds per week is 2.3534 standard deviations above the mean, making it higher than the average income.
Hint: zscore = (x - mean)/sd
income = mydata$income
income [1] 2.00 2.50 2.00 7.00 2.00 3.47 5.50 6.42
[9] 2.00 6.00 3.00 4.75 2.20 2.00 3.00 1.50
[17] 9.50 10.00 4.00 3.50 3.00 2.50 3.50 10.00
[25] 6.50 1.50 5.44 1.00 0.60 5.50 12.00 7.00
[33] 15.00 2.00 1.50 4.50 2.50 8.00 10.00 1.60
[41] 2.00 15.00 3.00 3.25 4.94 1.50 2.50zscoreincomeX = (13 - mean(income)) / sd(income)
zscoreincomeX[1] 2.353481Question 5: Create a histogram for the zscore of income. What do you notice about the shape? ( 5 Points )
The histogram for the z scores of income has its highest frequency at a z score that deviates from the mean by about 1 standard deviation. This means that it is skewed to the left. The rest of the z scores havae a much lower frequency as the z scores increase meaning less z scores were positive positioned away from the mean. Hint: To plot a histogram, use the function hist(variable).
zscoreincom = (income - mean(income)) / sd(income)
zscoreincom [1] -0.74391403 -0.60312335 -0.74391403 0.66399285
[5] -0.74391403 -0.32998941 0.24162079 0.50067565
[9] -0.74391403 0.38241147 -0.46233266 0.03043475
[13] -0.68759776 -0.74391403 -0.46233266 -0.88470472
[17] 1.36794630 1.50873698 -0.18075128 -0.32154197
[21] -0.46233266 -0.60312335 -0.32154197 1.50873698
[25] 0.52320216 -0.88470472 0.22472590 -1.02549541
[29] -1.13812796 0.24162079 2.07189974 0.66399285
[33] 2.91664387 -0.74391403 -0.88470472 -0.03996059
[37] -0.60312335 0.94557423 1.50873698 -0.85654659
[41] -0.74391403 2.91664387 -0.46233266 -0.39193731
[45] 0.08393521 -0.88470472 -0.60312335hist(zscoreincom)
Question 6: Analyze the correlation plot below. Give relavant information about the negative correlated, no correlared and positive correlated variables. ( 5 Points )
This correlation plot shows positive correlations, and lack of correlations becasue there are many differntly shaded blue dots, but no negative correlations becuase there are no red dots. Correlations range from -1 to 1. -1 is a strong negative correlation, 0 is no correlation, and 1 is a strong positive correlation. One of the stronger positive correlations is between fb_likes and user_votes which has a positive correlation of around .8. There is almost no correlation between score and cast_likes or cast_likes and director. There is also a fairly strong correlation of about positive .6 between gross and fb_likes.Score and cast likes has very little correlation, almost coming in at 0. This information is useful because you can see which two variable are related and use this to try to boost things like fb_likes or score.