AAQ2
Maisarah binti Maizul Lizan (17206792/1)
a) Confusion Matrix
- The model correctly predicted the positive class for COVID-19 screening test 120 times and incorrectly predicted it 15 times.
- The model correctly predicted the negative class for COVID-19 screening test 50 times and incorrectly predicted it 10 times.
- The model made 170 correct predictions (120+50).
- The model made 25 incorrect predictions (15+10).
- The total scored cases is 195.
- The error rate is 25/195 = 0.128.
- The overall accuracy rate is 170/195 = 0.8718.
b) Exploratory Data Analysis
Weight & Height of American women aged 30-36
data(women)
women## height weight
## 1 58 115
## 2 59 117
## 3 60 120
## 4 61 123
## 5 62 126
## 6 63 129
## 7 64 132
## 8 65 135
## 9 66 139
## 10 67 142
## 11 68 146
## 12 69 150
## 13 70 154
## 14 71 159
## 15 72 164Summary of the height of the sample of American women:
- Mean, median, min and max.
summary(women$height)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 58.0 61.5 65.0 65.0 68.5 72.0To describe the data and provide a summary, the summary() function is used on the height variable. Based on the result above, the minimum height is 58 inches. The maximum height is 72 inches. The mean is 65. The median is also 65.
- Quantile
quantile(women$height)## 0% 25% 50% 75% 100%
## 58.0 61.5 65.0 68.5 72.0To find the quantile of the height, the quartile() function is used.
- Interquartile range
IQR(women$height)## [1] 7To find the interquartile range of the height, the IQR() function is used. The interquartile range is 7.
- Variance
var(women$height)## [1] 20To find the variance of the height, the var() function is used. The variance is 20.
- Standard deviation
sd(women$height)## [1] 4.472136To find the standard deviation of the height, the sd() function is used. The standard deviation is 4.472136.
Summary of the weight of the sample of American women:
- Mean, median, min and max.
summary(women$weight)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 115.0 124.5 135.0 136.7 148.0 164.0To describe the data and provide a summary, the summary() function is used on the weight variable. Based on the result above, the minimum weight is 115 lbs. The maximum weight is 164 lbs. The mean is 135.0. The median is 136.7.
- Quantile
quantile(women$weight)## 0% 25% 50% 75% 100%
## 115.0 124.5 135.0 148.0 164.0To find the quantile of the height, the quantile() function is used.
- Interquartile Range
IQR(women$weight)## [1] 23.5To find the interquartile range of the height, the IQR() function is used. The interquartile range is 23.5.
- Variance
var(women$weight)## [1] 240.2095To find the variance of the height, the var() function is used. The variance is 240.2095.
- Standard Deviation
sd(women$weight)## [1] 15.49869To find the standard deviation of the height, the sd() function is used. The standard deviation is 15.49869.
Plot:
plot(women, xlab = "Height (in)", ylab = "Weight (lb)", main = "Weight & Height of American Women aged 30-39")
fit <- lm(weight~height, data = women)
abline(fit, lty="dashed")
Codebook:
codebook(women)## ================================================================================
##
## height
##
## --------------------------------------------------------------------------------
##
## Storage mode: double
##
## Min: 58.000
## Max: 72.000
## Mean: 65.000
## Std.Dev.: 4.320
## Skewness: 0.000
## Kurtosis: -1.211
##
## ================================================================================
##
## weight
##
## --------------------------------------------------------------------------------
##
## Storage mode: double
##
## Min: 115.000
## Max: 164.000
## Mean: 136.733
## Std.Dev.: 14.973
## Skewness: 0.252
## Kurtosis: -1.100summary(women)## height weight
## Min. :58.0 Min. :115.0
## 1st Qu.:61.5 1st Qu.:124.5
## Median :65.0 Median :135.0
## Mean :65.0 Mean :136.7
## 3rd Qu.:68.5 3rd Qu.:148.0
## Max. :72.0 Max. :164.0c) Data Manipulation with dplyr
states <- c("Perlis","Kedah","Penang","Perak","Selangor","Kuala Lumpur","Putrajaya", "Malacca","Johore","Kelantan","Terengganu", "Negeri Sembilan","Pahang")
region <- c("Northern","Northern","Northern","Northern","Central","Central","Central","Southern","Southern","East Coast","East Coast","Southern","East Coast")
cases <- c(8,422,372,279,2235,447,30,209,549,626,266,434,263)
covidmalaysia <- data.frame(states, cases, region)
print(covidmalaysia)## states cases region
## 1 Perlis 8 Northern
## 2 Kedah 422 Northern
## 3 Penang 372 Northern
## 4 Perak 279 Northern
## 5 Selangor 2235 Central
## 6 Kuala Lumpur 447 Central
## 7 Putrajaya 30 Central
## 8 Malacca 209 Southern
## 9 Johore 549 Southern
## 10 Kelantan 626 East Coast
## 11 Terengganu 266 East Coast
## 12 Negeri Sembilan 434 Southern
## 13 Pahang 263 East CoastThis dataset represents the number of Covid-19 cases in Peninsular Malaysia on 23 May 2021. The source is taken from the official website of Ministry of Health. The data is retrieved from covid-19.moh.gov.my
- Change the column name
covidmalaysia2 <- rename(covidmalaysia, c(Peninsular_States=states, Covid19_Cases = cases))
covidmalaysia2## Peninsular_States Covid19_Cases region
## 1 Perlis 8 Northern
## 2 Kedah 422 Northern
## 3 Penang 372 Northern
## 4 Perak 279 Northern
## 5 Selangor 2235 Central
## 6 Kuala Lumpur 447 Central
## 7 Putrajaya 30 Central
## 8 Malacca 209 Southern
## 9 Johore 549 Southern
## 10 Kelantan 626 East Coast
## 11 Terengganu 266 East Coast
## 12 Negeri Sembilan 434 Southern
## 13 Pahang 263 East CoastThe rename() function is used to change the column names. The column ‘states’ is changed to ‘Peninsular_States’. The column ‘cases’ is changed to ‘Covid19_Cases’.
- Pick rows based on their values
covidmalaysia2 %>% filter(region=="Central")## Peninsular_States Covid19_Cases region
## 1 Selangor 2235 Central
## 2 Kuala Lumpur 447 Central
## 3 Putrajaya 30 CentralThe filter()function is used to pick rows that contains the value “Central” from the dataset called “covidmalaysia2”. The output produced Kuala Lumpur, Selangor, and Putrajaya.
- Add new columns to a data frame
cluster <- c(0, 79, 121,20,72,40,0,59,194,140,113,55,117)
closecontact <- c(6,194,110,167,1768,263,23,99,254,361,112,253, 110)
covidmalaysia2 %>% mutate(cluster,closecontact, .after=Covid19_Cases)## Peninsular_States Covid19_Cases cluster closecontact region
## 1 Perlis 8 0 6 Northern
## 2 Kedah 422 79 194 Northern
## 3 Penang 372 121 110 Northern
## 4 Perak 279 20 167 Northern
## 5 Selangor 2235 72 1768 Central
## 6 Kuala Lumpur 447 40 263 Central
## 7 Putrajaya 30 0 23 Central
## 8 Malacca 209 59 99 Southern
## 9 Johore 549 194 254 Southern
## 10 Kelantan 626 140 361 East Coast
## 11 Terengganu 266 113 112 East Coast
## 12 Negeri Sembilan 434 55 253 Southern
## 13 Pahang 263 117 110 East CoastTwo new columns called ‘cluster’ and ‘closecontact’ is added to the dataset using the mutate() function. The data is retrieved from covid-19.moh.gov.my
- Combine dataset
covidmalaysia3 = data.frame(Peninsular_States = c("Perlis","Kedah","Penang","Perak","Selangor","Kuala Lumpur","Putrajaya", "Malacca","Johore","Kelantan","Terengganu", "Negeri Sembilan","Pahang"), importcase = c(0,0,0,0,0,4,0,0,0,0,0,0,0))
covidmalaysia4 = left_join(x=covidmalaysia2, y=covidmalaysia3, by="Peninsular_States")
covidmalaysia4## Peninsular_States Covid19_Cases region importcase
## 1 Perlis 8 Northern 0
## 2 Kedah 422 Northern 0
## 3 Penang 372 Northern 0
## 4 Perak 279 Northern 0
## 5 Selangor 2235 Central 0
## 6 Kuala Lumpur 447 Central 4
## 7 Putrajaya 30 Central 0
## 8 Malacca 209 Southern 0
## 9 Johore 549 Southern 0
## 10 Kelantan 626 East Coast 0
## 11 Terengganu 266 East Coast 0
## 12 Negeri Sembilan 434 Southern 0
## 13 Pahang 263 East Coast 0By using the left_join() function, a dataframe called ‘covidmalaysia3’ is combined with ‘covidmalaysia2’, and a dataset called covidmalaysia4 is created. The data is retrieved from covid-19.moh.gov.my