## P(Z>-1.13) = 1−P(Z<−1.13)
Z = -1.13
normalPlot(bounds=c(Z,Inf))
Z = 0.18
normalPlot(bounds=c(-Inf,Z))
Z = 8
normalPlot(bounds=c(Z,Inf))
Z = 0.5
normalPlot(bounds=c(-Z,Z))
Men (Age:30-34): N(μ=4313,σ=583)
Women (Age:25-29): N(μ=5261,σ=807)
Leo:
Z_leo =(4948-4313) / 583
Z_mary =(5513-5261) / 807
Z_leo
## [1] 1.089194
Z_mary
## [1] 0.3122677
The Z-score transforms individual’s performance into standard normal distribution and allow us to compare them on a sam scale. According to the Z-score, Leo has worse performance than Mary.
In this case, we are comparing the finishing time of competitors. Thus, the more left the performance positioning in the normal distribution, the better the performance. Mary’s Z-score is 0.31 which is to the left of Leo’s Z-score. Thus Mary’s performance is better than Leo’s performance.
About 13.8% pf triathletes did Leo finish faster than in his group.
normalPlot(bounds=c(Z_leo,Inf))
About 37.7% pf triathletes did Leo finish faster than in his group.
normalPlot(bounds=c(Z_mary,Inf))
If the distribution of finishing time are not nearly normal, we cannot use standardization to find Z-Score for each finisher.
hgt = c(54, 55, 56, 56, 57, 58, 58, 59, 60, 60, 60, 61, 61, 62, 62, 63, 63, 63, 64, 65, 65, 67, 67, 69, 73)
mean = mean(hgt)
sd = sd(hgt)
for (n in c(1,2,3)){
sd_n= hgt[hgt>mean-sd*n & hgt<mean+sd*n]
perc=length(sd_n) / length(hgt)
print(perc)
}
## [1] 0.68
## [1] 0.96
## [1] 1
Given the proportion of 68%, 96%, 100% for respectively 1,2,3* standard deviations, the distribution of heights does approximately follow the rule.
Most part of the data dp appear to follow a normal distribution.
qqnormsim(hgt)
This case fits a geomtric distribution.
df_rate = 0.02
n = 9
dgeom(n, df_rate)
## [1] 0.01667496
The probability is 13.26%
(1-0.02)^100
## [1] 0.1326196
On average you should expect 50 transistors to be produced before the first with a defect. The standad deviation is 0.1443.
mean =1/df_rate
sd = sqrt((df_rate)/(1-df_rate)^2)
mean
## [1] 50
sd
## [1] 0.1443075
On average you should expect 20 transistors to be produced before the first with a defect. The standad deviation is 0.2354.
df_rate = 0.05
mean =1/df_rate
sd = sqrt((df_rate)/(1-df_rate)^2)
mean
## [1] 20
sd
## [1] 0.2353756
The increase of probability of an even will make waite time untill success shorter. Will make the sd greater.
The probability that two of them will be boys is 38.23%
p_boy = 0.51
dbinom(2, 3,p_boy)
## [1] 0.382347
The possible sequences are : bbg, bgb, gbb
p_boy^2 * (1-p_boy)*3
## [1] 0.382347
The answers match.
Using part b, you will have to list out all possible sequence, which is very tedious. The probability is 20.98%
dbinom(3, 8,p_boy)
## [1] 0.2098355
Given the 3rd success serve has to be the 10th trail, it could be interprete as the first 9 trial has 2 success server and the 10th trial is success.
p_serve = 0.15
dbinom(2,9,p_serve)*p_serve
## [1] 0.03895012
The probability is 3.9%.
Given the assumption that her serves are independent of each other, the 10th serve being successful is still 15%.
Part a and part b calculates different events’ probability. Part A sees the 10 events as a whole while part b only see the 10th events. Thus the calculation should be different.