R Bridge Course Final Project
Amit Kapoor
03 August 2019
Data Description
Births data frame has 7305 observations on the following 8 columns.
- date - Date
- births - Number of births on date (integer)
- wday - Day of week (ordered factor)
- year - Year (integer)
- month - Month (integer)
- day_of_year - Day of year (integer)
- day_of_month - Day of month (integer)
- day_of_week - Day of week (integer)
library(psych)
library(ggplot2)##
## Attaching package: 'ggplot2'## The following objects are masked from 'package:psych':
##
## %+%, alpha1. Data Exploration: This should include summary statistics, means, medians, quartiles, or any other relevant information about the data set. Please include some conclusions in the R Markdown text.
I will target below questions.
- Which year has maximum and minimum number of births births?
- Which month has maximum and minimum number of births births?
- How is the birth rate being for the given years in data?
- Is there a pattern on number of births?
theLink <- "https://raw.githubusercontent.com/amit-kapoor/r/master/data/births.csv"
# load data into data frame
df_births <- read.csv(file=theLink, header = TRUE, sep = ",")
# display header rows
head(df_births)## X date births wday year month day_of_year day_of_month day_of_week
## 1 1 1969-01-01 8486 Wed 1969 1 1 1 3
## 2 2 1969-01-02 9002 Thu 1969 1 2 2 4
## 3 3 1969-01-03 9542 Fri 1969 1 3 3 5
## 4 4 1969-01-04 8960 Sat 1969 1 4 4 6
## 5 5 1969-01-05 8390 Sun 1969 1 5 5 7
## 6 6 1969-01-06 9560 Mon 1969 1 6 6 1# summary of Birth data
summary(df_births)## X date births wday
## Min. : 1 1969-01-01: 1 Min. : 6675 Fri:1044
## 1st Qu.:1827 1969-01-02: 1 1st Qu.: 8792 Mon:1043
## Median :3653 1969-01-03: 1 Median : 9622 Sat:1044
## Mean :3653 1969-01-04: 1 Mean : 9649 Sun:1043
## 3rd Qu.:5479 1969-01-05: 1 3rd Qu.:10510 Thu:1044
## Max. :7305 1969-01-06: 1 Max. :12851 Tue:1043
## (Other) :7299 Wed:1044
## year month day_of_year day_of_month
## Min. :1969 Min. : 1.000 Min. : 1.0 Min. : 1.00
## 1st Qu.:1974 1st Qu.: 4.000 1st Qu.: 93.0 1st Qu.: 8.00
## Median :1979 Median : 7.000 Median :184.0 Median :16.00
## Mean :1979 Mean : 6.523 Mean :183.8 Mean :15.73
## 3rd Qu.:1984 3rd Qu.:10.000 3rd Qu.:275.0 3rd Qu.:23.00
## Max. :1988 Max. :12.000 Max. :366.0 Max. :31.00
##
## day_of_week
## Min. :1
## 1st Qu.:2
## Median :4
## Mean :4
## 3rd Qu.:6
## Max. :7
## # describe Birth data
describe(df_births)## vars n mean sd median trimmed mad min max
## X 1 7305 3653.00 2108.92 3653 3653.00 2707.23 1 7305
## date* 2 7305 3653.00 2108.92 3653 3653.00 2707.23 1 7305
## births 3 7305 9648.94 1127.32 9622 9648.80 1275.04 6675 12851
## wday* 4 7305 4.00 2.00 4 4.00 2.97 1 7
## year 5 7305 1978.50 5.77 1979 1978.50 7.41 1969 1988
## month 6 7305 6.52 3.45 7 6.53 4.45 1 12
## day_of_year 7 7305 183.75 105.62 184 183.82 134.92 1 366
## day_of_month 8 7305 15.73 8.80 16 15.72 11.86 1 31
## day_of_week 9 7305 4.00 2.00 4 4.00 2.97 1 7
## range skew kurtosis se
## X 7304 0.00 -1.20 24.67
## date* 7304 0.00 -1.20 24.67
## births 6176 0.02 -0.67 13.19
## wday* 6 0.00 -1.25 0.02
## year 19 0.00 -1.21 0.07
## month 11 -0.01 -1.21 0.04
## day_of_year 365 0.00 -1.20 1.24
## day_of_month 30 0.01 -1.19 0.10
## day_of_week 6 0.00 -1.25 0.022. Data wrangling: Please perform some basic transformations. They will need to make sense but could include column renaming, creating a subset of the data, replacing values, or creating new columns with derived data (for example – if it makes sense you could sum two columns together)
# unique values of month column
unique(df_births$month)## [1] 1 2 3 4 5 6 7 8 9 10 11 12# replace months numeric value to corressponding month name
df_births$month[df_births$month == 1] <- "Jan"
df_births$month[df_births$month == 2] <- "Feb"
df_births$month[df_births$month == 3] <- "Mar"
df_births$month[df_births$month == 4] <- "Apr"
df_births$month[df_births$month == 5] <- "May"
df_births$month[df_births$month == 6] <- "Jun"
df_births$month[df_births$month == 7] <- "Jul"
df_births$month[df_births$month == 8] <- "Aug"
df_births$month[df_births$month == 9] <- "Sep"
df_births$month[df_births$month == 10] <- "Oct"
df_births$month[df_births$month == 11] <- "Nov"
df_births$month[df_births$month == 12] <- "Dec"
head(df_births)## X date births wday year month day_of_year day_of_month day_of_week
## 1 1 1969-01-01 8486 Wed 1969 Jan 1 1 3
## 2 2 1969-01-02 9002 Thu 1969 Jan 2 2 4
## 3 3 1969-01-03 9542 Fri 1969 Jan 3 3 5
## 4 4 1969-01-04 8960 Sat 1969 Jan 4 4 6
## 5 5 1969-01-05 8390 Sun 1969 Jan 5 5 7
## 6 6 1969-01-06 9560 Mon 1969 Jan 6 6 1# unique values after replacement
unique(df_births$month)## [1] "Jan" "Feb" "Mar" "Apr" "May" "Jun" "Jul" "Aug" "Sep" "Oct" "Nov"
## [12] "Dec"# maximum value in births columns
max(df_births$births)## [1] 12851# minimum value in births columns
min(df_births$births)## [1] 6675# group by births by year
df_tot_birth_year = aggregate(df_births$births, by=list(Category=df_births$year), FUN=sum)
df_tot_birth_year## Category x
## 1 1969 3598578
## 2 1970 3734606
## 3 1971 3563074
## 4 1972 3265863
## 5 1973 3145670
## 6 1974 3170306
## 7 1975 3153366
## 8 1976 3176266
## 9 1977 3332048
## 10 1978 3338197
## 11 1979 3499737
## 12 1980 3617948
## 13 1981 3635480
## 14 1982 3685374
## 15 1983 3642731
## 16 1984 3673531
## 17 1985 3765054
## 18 1986 3760682
## 19 1987 3813211
## 20 1988 3913786# year having maximum number of births
df_tot_birth_year$Category[df_tot_birth_year$x == max(df_tot_birth_year$x)]## [1] 1988# year having minimum number of births
df_tot_birth_year$Category[df_tot_birth_year$x == min(df_tot_birth_year$x)]## [1] 1973# group by births by month
df_tot_birth_month = aggregate(df_births$births, by=list(Category=df_births$month), FUN=sum)
df_tot_birth_month## Category x
## 1 Apr 5560775
## 2 Aug 6309764
## 3 Dec 5904376
## 4 Feb 5362585
## 5 Jan 5759165
## 6 Jul 6220295
## 7 Jun 5758571
## 8 Mar 5868140
## 9 May 5785220
## 10 Nov 5695273
## 11 Oct 6055372
## 12 Sep 6205972# month having maximum number of births
df_tot_birth_month$Category[df_tot_birth_month$x == max(df_tot_birth_month$x)]## [1] "Aug"# month having minimum number of births
df_tot_birth_month$Category[df_tot_birth_month$x == min(df_tot_birth_month$x)]## [1] "Feb"3. Graphics: Please make sure to display at least one scatter plot, box plot and histogram. Don’t be limited to this. Please explore the many other options in R packages such as ggplot2.
# plot births vs year
plot(x = df_births$births, y=df_births$year, xlab = "Births", ylab = "Year", main = "Births vs Year", col = 'red')
ggplot(df_births, aes(x=df_births$year, y=df_births$births)) + geom_point()
# histogram for births by frequency
hist(df_births$births, freq = TRUE, col="green")
# histogram for births by density
hist(df_births$births, freq = FALSE, col="blue")
# boxplot births and year
boxplot(df_births$births ~ df_births$year, col = "orange")
# boxplot births and month
boxplot(df_births$births ~ df_births$month, col = "yellow")
4. Meaningful question for analysis: Please state at the beginning a meaningful question for analysis. Use the first three steps and anything else that would be helpful to answer the question you are posing from the data set you chose. Please write a brief conclusion paragraph in R markdown at the end.
- year having maximum number of births - 1988
- year having minimum number of births - 1973
- month having maximum number of births - Aug
- month having minimum number of births - Feb
- plots shown in section 3, shows the births data for given years.
- histogram above shows the distribution of births data.
Conclusion
I have looked into births data for years and months available in data. I do see a very close normal distribution trend in births data above which doesnt seem much skewed.