Problem Set # 3
Jenine Massat
date()## [1] "Tue Oct 03 19:56:59 2017"Due Date: September 27, 2017 Total Points: 45
1 The dataset hflights from the hflights package contains all 227,496 flights that departed Houston in 2011. Using the functions in the dplyr package
- Create a new data frame from hflights containing only those flights that departed on September 11th of that year. (5)
library(hflights)
library(dplyr)## Warning: package 'dplyr' was built under R version 3.4.2##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(ggplot2)
hflights.df = tbl_df(hflights)sep11.df = hflights %>%
filter(Month == "9" & DayofMonth == "11")
head(sep11.df)## Year Month DayofMonth DayOfWeek DepTime ArrTime UniqueCarrier FlightNum
## 1 2011 9 11 7 1546 1651 AA 458
## 2 2011 9 11 7 551 904 AA 466
## 3 2011 9 11 7 1936 2036 AA 657
## 4 2011 9 11 7 1438 1544 AA 742
## 5 2011 9 11 7 1720 2030 AA 1294
## 6 2011 9 11 7 1142 1258 AA 1848
## TailNum ActualElapsedTime AirTime ArrDelay DepDelay Origin Dest Distance
## 1 N559AA 65 40 -14 -4 IAH DFW 224
## 2 N3EGAA 133 115 -16 -9 IAH MIA 964
## 3 N498AA 60 40 -19 -4 IAH DFW 224
## 4 N470AA 66 43 9 18 IAH DFW 224
## 5 N3BVAA 130 118 -20 -5 IAH MIA 964
## 6 N598AA 76 40 -2 -3 IAH DFW 224
## TaxiIn TaxiOut Cancelled CancellationCode Diverted
## 1 12 13 0 0
## 2 5 13 0 0
## 3 8 12 0 0
## 4 6 17 0 0
## 5 5 7 0 0
## 6 22 14 0 0- How many flights departed on that day? (5)
taxiout.df = sep11.df %>%
select(TaxiOut)
head(taxiout.df)## TaxiOut
## 1 13
## 2 13
## 3 12
## 4 17
## 5 7
## 6 14sum(taxiout.df)## [1] 80118011 were taxied out or departed.
- Create a data frame with the first column being the tail number and the second being the number of departures from Houston the plane made that year sorted by most to least number of flights. (5)
tailanddepart.df = hflights %>%
select(TailNum, TaxiOut) %>%
arrange(desc(TaxiOut))
head(tailanddepart.df)## TailNum TaxiOut
## 1 N14998 163
## 2 N27190 152
## 3 N29917 151
## 4 N14558 151
## 5 N510SW 150
## 6 N31412 1462 Answer the following questions using the tornado data set (http://myweb.fsu.edu/jelsner/temp/data/Tornadoes.txt).
L = "http://myweb.fsu.edu/jelsner/temp/data/Tornadoes.txt"
Torn.df = read.table(L, header = TRUE)
head(Torn.df)## OM YEAR MONTH DAY DATE TIME TIMEZONE STATE FIPS STATENUMBE FSCALE
## 1 1 1950 1 3 1950-01-03 1100 3 MO 29 1 3
## 2 2 1950 1 3 1950-01-03 1155 3 IL 17 2 3
## 3 3 1950 1 3 1950-01-03 1600 3 OH 39 1 1
## 4 4 1950 1 13 1950-01-13 525 3 AR 5 1 3
## 5 5 1950 1 25 1950-01-25 1930 3 MO 29 2 2
## 6 6 1950 1 25 1950-01-25 2100 3 IL 17 3 2
## INJURIES FATALITIES LOSS CROPLOSS SLAT SLON ELAT ELON LENGTH WIDTH
## 1 3 0 6 0 38.77 -90.22 38.83 -90.03 9.5 150
## 2 3 0 5 0 39.10 -89.30 39.12 -89.23 3.6 130
## 3 1 0 4 0 40.88 -84.58 0.00 0.00 0.1 10
## 4 1 1 3 0 34.40 -94.37 0.00 0.00 0.6 17
## 5 5 0 5 0 37.60 -90.68 37.63 -90.65 2.3 300
## 6 0 0 5 0 41.17 -87.33 0.00 0.00 0.1 100
## NS SN SG F1 F2 F3 F4
## 1 2 0 1 0 0 0 0
## 2 1 1 1 135 0 0 0
## 3 1 1 1 161 0 0 0
## 4 1 1 1 113 0 0 0
## 5 1 1 1 93 0 0 0
## 6 1 1 1 91 0 0 0Torn.df = tbl_df(Torn.df)
Torn.df## # A tibble: 56,221 x 28
## OM YEAR MONTH DAY DATE TIME TIMEZONE STATE FIPS
## * <int> <int> <int> <int> <fctr> <int> <int> <fctr> <int>
## 1 1 1950 1 3 1950-01-03 1100 3 MO 29
## 2 2 1950 1 3 1950-01-03 1155 3 IL 17
## 3 3 1950 1 3 1950-01-03 1600 3 OH 39
## 4 4 1950 1 13 1950-01-13 525 3 AR 5
## 5 5 1950 1 25 1950-01-25 1930 3 MO 29
## 6 6 1950 1 25 1950-01-25 2100 3 IL 17
## 7 7 1950 1 26 1950-01-26 1800 3 TX 48
## 8 8 1950 2 11 1950-02-11 1310 3 TX 48
## 9 9 1950 2 11 1950-02-11 1350 3 TX 48
## 10 10 1950 2 11 1950-02-11 2100 3 TX 48
## # ... with 56,211 more rows, and 19 more variables: STATENUMBE <int>,
## # FSCALE <int>, INJURIES <int>, FATALITIES <int>, LOSS <int>,
## # CROPLOSS <int>, SLAT <dbl>, SLON <dbl>, ELAT <dbl>, ELON <dbl>,
## # LENGTH <dbl>, WIDTH <int>, NS <int>, SN <int>, SG <int>, F1 <int>,
## # F2 <int>, F3 <int>, F4 <int>- Use the functions
group_by()andsummarize()to determine the number of tornadoes by each state. (5)
Torn.df %>%
group_by(STATE) %>%
summarize(STATENUMBE = max(STATENUMBE, na.rm = TRUE))## # A tibble: 52 x 2
## STATE STATENUMBE
## <fctr> <dbl>
## 1 AK 2
## 2 AL 145
## 3 AR 105
## 4 AZ 17
## 5 CA 30
## 6 CO 98
## 7 CT 8
## 8 DC 1
## 9 DE 6
## 10 FL 115
## # ... with 42 more rows- Create a new column with path length in meters. Create a new data frame by removing rows with EF damage rating below 3. Group by year and summarize the average path length. Make a graph with year on the x axis and average path length on the y axis. (10)
EFless3.df = Torn.df %>%
mutate(Length = LENGTH * 1609) %>%
filter(FSCALE > 3) %>%
select(OM:SG,F3:F4)
EFless3.df## # A tibble: 594 x 26
## OM YEAR MONTH DAY DATE TIME TIMEZONE STATE FIPS
## <int> <int> <int> <int> <fctr> <int> <int> <fctr> <int>
## 1 20 1950 2 12 1950-02-12 1300 3 LA 22
## 2 59 1950 4 28 1950-04-28 1800 3 TX 48
## 3 60 1950 4 28 1950-04-28 1905 3 OK 40
## 4 79 1950 5 4 1950-05-04 2310 3 KS 20
## 5 132 1950 6 8 1950-06-08 2010 3 KS 20
## 6 150 1950 6 25 1950-06-25 2100 3 WI 55
## 7 164 1950 7 15 1950-07-15 1730 3 NE 31
## 8 66 1951 5 18 1951-05-18 1545 3 TX 48
## 9 149 1951 6 19 1951-06-19 1730 3 MN 27
## 10 171 1951 6 27 1951-06-27 10 3 KS 20
## # ... with 584 more rows, and 17 more variables: STATENUMBE <int>,
## # FSCALE <int>, INJURIES <int>, FATALITIES <int>, LOSS <int>,
## # CROPLOSS <int>, SLAT <dbl>, SLON <dbl>, ELAT <dbl>, ELON <dbl>,
## # LENGTH <dbl>, WIDTH <int>, NS <int>, SN <int>, SG <int>, F3 <int>,
## # F4 <int>mlyear.df = EFless3.df %>%
group_by(YEAR) %>%
summarise(mL = mean(LENGTH))
mlyear.df## # A tibble: 62 x 2
## YEAR mL
## <int> <dbl>
## 1 1950 23.27143
## 2 1951 18.16000
## 3 1952 16.79444
## 4 1953 27.43182
## 5 1954 38.05714
## 6 1955 36.57000
## 7 1956 42.01538
## 8 1957 28.09231
## 9 1958 22.62000
## 10 1959 18.14286
## # ... with 52 more rowsmlyear.df %>%
ggplot(aes(YEAR, mL))+
geom_point()
c. Compute the number of tornadoes with EF damage rating 1 or higher by year and graph the results. (10)
Torn.df %>%
filter(FSCALE >= 1)%>%
select(YEAR,FSCALE) %>%
ggplot(aes(YEAR, FSCALE))+
geom_bar(stat = "identity")
- Create a data frame with the year in the first column and the total number of tornadoes in Kansas by year in the second column. List the first six rows of this new data frame. (5)
KSTor.df = Torn.df %>%
select(YEAR,STATE,STATENUMBE) %>%
group_by(STATE) %>%
filter(STATE == "KS")
head(KSTor.df)## # A tibble: 6 x 3
## # Groups: STATE [1]
## YEAR STATE STATENUMBE
## <int> <fctr> <int>
## 1 1950 KS 1
## 2 1950 KS 2
## 3 1950 KS 3
## 4 1950 KS 4
## 5 1950 KS 5
## 6 1950 KS 6