An easy tool for visualizing, annotating, and stratifying time series data.
This package contains five functions as below.
timelineS
: Plots a horizontal timeline with event
descriptions at corresponding dates.timelineG
: Plots faceted timelines for grouped
data.durPlot
: Plots boxplot, histogram, density plot,
scatter plot, line plot and prints summary statistics for date duration
data.durCalc
: Calculates the duration between two dates, use
it as a filter to select rows that satisfy the length criteria. Returns
the dataset with additional columns regarding the length of durations in
different units.durSummary
: Returns summary statistics for date
duration data.install.packages("timelineS")
::install_github("daheelee/timelineS") devtools
timelineS
plots an annotated timeline.
timelineS(mj_life, main = "Life of Michael Jackson")
You can also change the aesthetics.
timelineS(mj_life, main = "Life of Michael Jackson",
label.direction = "up", label.length = c(0.2,0.8,0.4,1.2), label.position = 3,
line.color = "blue", label.color = "blue", point.color = "blue", pch = "-")
timelineG
creates stacked timelines, faceted by
groups.
timelineG(df=life_country, start="Start", end="End", names="Name",
phase="Phase", group1="Country", group2="Gender")
durPlot
function gives five different plots by default.
You can set facet=TRUE
to get faceted plots.
durPlot(life_exp, start="Birth", end="Death", group="Country", timeunit="years", facet=TRUE, binwidth=3, alpha=0.7, title=TRUE)
durPlot
also gives a summary of the duration lengths
with the plots.
durSummary
is a simpler function if you only want the
summary.
durSummary(life_exp, start="Birth", end="Death", group="Country", timeunit="years")
Country min Qt1 median mean Qt3 max sd1 Australia 67.19 72.16 76.26 75.72 78.98 85.71 5.30
2 China 67.41 70.09 76.06 76.44 83.38 87.56 7.14
3 France 64.02 68.16 74.48 74.89 79.10 89.37 7.59
4 Japan 69.35 75.14 77.76 77.63 80.57 85.33 4.82
5 US 62.73 72.04 76.21 75.35 79.00 85.45 5.49
durCalc
filters a data frame using duration lengths.
Let’s say we have a dataset that has dates of birth and death of people.
We can filter out people(rows) who lived longer than 85 years. You can
also do filterlonger=FALSE
to select rows that are shorter
than some time length.
durCalc(life_exp, start="Birth", end="Death", timeunit="years", filterlength=85)
The output has a few extra duration information in specified
timeunit
and also in calendar units.
Name Country Gender Birth Death days diff_years diff_length longer_by8 Susan Australia Female 1920-05-26 2006-02-10 31306 85.71 85years 8months 16days 8months 16days
19 Mai Japan Female 1923-10-12 2009-02-07 31165 85.33 85years 3months 27days 3months 27days
24 Leo France Male 1920-06-28 2009-11-09 32641 89.37 89years 4months 12days 4years 4months 12days
48 Gloria US Female 1926-03-01 2011-08-14 31212 85.45 85years 5months 14days 5months 14days
60 Ming China Female 1920-09-06 2008-03-31 31983 87.56 87years 6months 24days 2years 6months 24days
If you want to know how old each person would be as of January 1, 2000, you can do:
durCalc(life_exp, start="Birth", end=as.Date("2000-1-1"), timeunit="years")
Name Country Gender Birth Death days diff_length1 Edward Australia Male 1927-11-17 1999-12-16 26343 72years 1month 15days
2 James Australia Male 1925-11-20 1993-01-27 27070 74years 1month 11days
3 Mark Australia Male 1926-06-11 2003-12-26 26867 73years 6months 21days
4 Fred Australia Male 1927-01-22 2002-07-19 26642 72years 11months 9days
5 Phil Australia Male 1923-06-20 2003-01-15 27954 76years 6months 12days
...
Or you can simply use this as a unit-converter between two dates.
durCalc(start=as.Date("2010-12-1"), end=as.Date("2015-4-26"), timeunit="weeks")
days diff_weeks diff_length1 1607 229.57 4years 4months 24days
Dahee Lee – dhlee99@gmail.com
Distributed under the GPL-2 license.