library(dplyr)
Attaching package: 'dplyr'The following objects are masked from 'package:stats':
filter, lagThe following objects are masked from 'package:base':
intersect, setdiff, setequal, unionlibrary(tidyr)
library(slider)Seattle_vs_NY
Approach
Use the data found using ChatGPT about weather in Seattle and New York. The data includes date and daily weather measurements such as maximum temperature, minimum temperature, precipitation, and wind. I will use slider package to get the six-day moving average
Code base
weather <- read.csv("C:/Users/CARES/Rstudio/DATA 607/weather.csv")head(weather) location date precipitation temp_max temp_min wind weather
1 Seattle 2012-01-01 0.0 12.8 5.0 4.7 drizzle
2 Seattle 2012-01-02 10.9 10.6 2.8 4.5 rain
3 Seattle 2012-01-03 0.8 11.7 7.2 2.3 rain
4 Seattle 2012-01-04 20.3 12.2 5.6 4.7 rain
5 Seattle 2012-01-05 1.3 8.9 2.8 6.1 rain
6 Seattle 2012-01-06 2.5 4.4 2.2 2.2 rainnames(weather)[1] "location" "date" "precipitation" "temp_max"
[5] "temp_min" "wind" "weather" Year to date average
weather$date <- as.Date(weather$date)
class(weather$date)[1] "Date"weather <- weather %>%
group_by(location, year = format(date, "%Y")) %>%
mutate(ytd_average = cummean(temp_max)) %>%
ungroup()weather %>%
filter(location == "Seattle") %>%
select(location, date, temp_max, ytd_average)# A tibble: 1,461 × 4
location date temp_max ytd_average
<chr> <date> <dbl> <dbl>
1 Seattle 2012-01-01 12.8 12.8
2 Seattle 2012-01-02 10.6 11.7
3 Seattle 2012-01-03 11.7 11.7
4 Seattle 2012-01-04 12.2 11.8
5 Seattle 2012-01-05 8.9 11.2
6 Seattle 2012-01-06 4.4 10.1
7 Seattle 2012-01-07 7.2 9.69
8 Seattle 2012-01-08 10 9.72
9 Seattle 2012-01-09 9.4 9.69
10 Seattle 2012-01-10 6.1 9.33
# ℹ 1,451 more rowsweather %>%
filter(location == "New York") %>%
select(location, date, temp_max, ytd_average)# A tibble: 1,461 × 4
location date temp_max ytd_average
<chr> <date> <dbl> <dbl>
1 New York 2012-01-01 10 10
2 New York 2012-01-02 10 10
3 New York 2012-01-03 0.6 6.87
4 New York 2012-01-04 -1.7 4.73
5 New York 2012-01-05 5.6 4.9
6 New York 2012-01-06 12.2 6.12
7 New York 2012-01-07 16.1 7.54
8 New York 2012-01-08 8.9 7.71
9 New York 2012-01-09 3.9 7.29
10 New York 2012-01-10 8.9 7.45
# ℹ 1,451 more rowsexplaination
The table shows the daily maximum temperatures in New York and the year-to-date average for 2012. The temp_max column shows the highest temperature each day. The ytd_average column shows the average of the temperatures from January 1 up to that date. For example, on January 3rd, the maximum temperature was 0.6 degrees, and the average temperature for the first three days was about 6.87 degrees. Compared to Seattle, the max temperature was 11.7 degrees on January 3rd. Seattle has an average of higher temperature than New York in January.
Next, six-day moving average
weather <- weather %>%
group_by(location) %>%
arrange(date) %>%
mutate(
six_day_average = slide_dbl(
temp_max,
mean,
.before = 5,
.complete = TRUE
)
) %>%
ungroup()weather %>%
select(date, location, temp_max, ytd_average, six_day_average) %>%
head(15)# A tibble: 15 × 5
date location temp_max ytd_average six_day_average
<date> <chr> <dbl> <dbl> <dbl>
1 2012-01-01 Seattle 12.8 12.8 NA
2 2012-01-01 New York 10 10 NA
3 2012-01-02 Seattle 10.6 11.7 NA
4 2012-01-02 New York 10 10 NA
5 2012-01-03 Seattle 11.7 11.7 NA
6 2012-01-03 New York 0.6 6.87 NA
7 2012-01-04 Seattle 12.2 11.8 NA
8 2012-01-04 New York -1.7 4.73 NA
9 2012-01-05 Seattle 8.9 11.2 NA
10 2012-01-05 New York 5.6 4.9 NA
11 2012-01-06 Seattle 4.4 10.1 10.1
12 2012-01-06 New York 12.2 6.12 6.12
13 2012-01-07 Seattle 7.2 9.69 9.17
14 2012-01-07 New York 16.1 7.54 7.13
15 2012-01-08 Seattle 10 9.72 9.07weather %>%
filter(location == "New York") %>%
select(location, date, six_day_average)# A tibble: 1,461 × 3
location date six_day_average
<chr> <date> <dbl>
1 New York 2012-01-01 NA
2 New York 2012-01-02 NA
3 New York 2012-01-03 NA
4 New York 2012-01-04 NA
5 New York 2012-01-05 NA
6 New York 2012-01-06 6.12
7 New York 2012-01-07 7.13
8 New York 2012-01-08 6.95
9 New York 2012-01-09 7.5
10 New York 2012-01-10 9.27
# ℹ 1,451 more rowsweather %>%
filter(location == "Seattle") %>%
select(location, date, six_day_average)# A tibble: 1,461 × 3
location date six_day_average
<chr> <date> <dbl>
1 Seattle 2012-01-01 NA
2 Seattle 2012-01-02 NA
3 Seattle 2012-01-03 NA
4 Seattle 2012-01-04 NA
5 Seattle 2012-01-05 NA
6 Seattle 2012-01-06 10.1
7 Seattle 2012-01-07 9.17
8 Seattle 2012-01-08 9.07
9 Seattle 2012-01-09 8.68
10 Seattle 2012-01-10 7.67
# ℹ 1,451 more rowsThis two tables above show New York and Seattle’s six-day moving average of maximum temperatures. The first five days display NA because there are not yet six days of temperature data available to calculate the average. Starting on January 6, the average is calculated using the current day and the previous five days. For example, on January 6, the six-day average is 10.1 in Seattle and 6.11 in New York.