Related
I have data on employment in different skill classes for different provinces over time. I'd like to show the employment over time in different provinces and classes in one graph. The following figure shows what I want but only for one year (2000)
ggplot(df, aes(fill=classes, y=total/10^6, x=province)) +
geom_bar(position="stack", stat="identity")
But I'd like to have each bar (showing each year) being repeated two times (in the example I have 2 years) for each province in the same graph. In other words, I'd like to show the 2001 data in the same graph as shown above beside the bars for 2000.
Here is part of the data:
df <- structure(list(year = c(2000L, 2000L, 2000L, 2000L, 2000L, 2000L,
2000L, 2000L, 2001L, 2001L, 2001L, 2001L, 2001L, 2001L, 2001L,
2001L), province = c("Alberta", "Alberta", "Alberta", "Alberta",
"Manitoba", "Manitoba", "Manitoba", "Manitoba", "Alberta", "Alberta",
"Alberta", "Alberta", "Manitoba", "Manitoba", "Manitoba", "Manitoba"
), classes = structure(c(2L, 3L, 4L, 1L, 2L, 3L, 4L, 1L, 2L,
3L, 4L, 1L, 2L, 3L, 4L, 1L), .Label = c("[0,0.2]", "(0.2,0.4]",
"(0.4,0.6]", "(0.6,0.8]", "(0.8,1)", "1"), class = "factor"),
total = c(11387250L, 4373500L, 18250L, 3215500L, 3984750L,
1414750L, 2000L, 1222750L, 11838250L, 4390000L, 21250L, 3272750L,
4019750L, 1331750L, 7750L, 1237000L)), row.names = c(NA,
-16L), vars = c("year", "province", "classes"), drop = TRUE, class = c("grouped_df",
"tbl_df", "tbl", "data.frame"), indices = list(3L, 0L, 1L, 2L,
7L, 4L, 5L, 6L, 11L, 8L, 9L, 10L, 15L, 12L, 13L, 14L), group_sizes = c(1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), biggest_group_size = 1L, labels = structure(list(
year = c(2000L, 2000L, 2000L, 2000L, 2000L, 2000L, 2000L,
2000L, 2001L, 2001L, 2001L, 2001L, 2001L, 2001L, 2001L, 2001L
), province = c("Alberta", "Alberta", "Alberta", "Alberta",
"Manitoba", "Manitoba", "Manitoba", "Manitoba", "Alberta",
"Alberta", "Alberta", "Alberta", "Manitoba", "Manitoba",
"Manitoba", "Manitoba"), classes = structure(c(1L, 2L, 3L,
4L, 1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L), .Label = c("[0,0.2]",
"(0.2,0.4]", "(0.4,0.6]", "(0.6,0.8]", "(0.8,1)", "1"), class = "factor")), row.names = c(NA,
-16L), vars = c("year", "province", "classes"), drop = TRUE, class = "data.frame"))
Is it what you are looking for ?
ggplot(df, aes(fill=classes, y=total/10^6, x=as.factor(year))) +
geom_bar(position="stack", stat="identity") +
facet_wrap(.~province)
As suggested by #user12728748, you can modify margin of the panel to make it more looks like a single plot:
ggplot(df, aes(fill=classes, y=total/10^6, x=as.factor(year))) +
geom_bar(position="stack", stat="identity") +
facet_wrap(.~province)+
theme(panel.margin = grid::unit(-1.25, "lines"))
NB: Be cautious because this trick can't be used if you set scales = free in your facet_wrap.
I have a data on countries and want to summarize it and create a table.
> head(data)
country year score members
A 1989 0 7
A 1990 0 7
A 1991 0 7
A 1992 0 7
A 1993 0 7
A 1994 0 7
The table should show the relationship between country "score" and the number of "members" – put differently, I want to see how many states with score 0,1 or 2 have "members"(ranging from 1 to 7).
I want to set it like this:
score members==1 members==2 members==3 members==4 members==5 members==6 members==7
0 1 0
1 2 0
2 0 1 and so on..
To do this I run the following:
library(dplyr)
table <- data %>%
group_by(score) %>%
summarise(
m1 = sum(members==1, na.rm=TRUE),
m2 = sum(members==2, na.rm=TRUE),
m3 = sum(members==3, na.rm=TRUE),
m4 = sum(members==4, na.rm=TRUE),
m5 = sum(members==5, na.rm=TRUE),
m6 = sum(members==6, na.rm=TRUE),
m7 = sum(members==7, na.rm=TRUE)
)
This gives:
score m1 m2 m3 m4 m5 m6 m7
0 0 2 0 0 0 3 30
1 15 3 11 11 3 18 3
2 3 0 2 2 0 6 9
.
.
I need a little help here. As you see it has calculated the total number of observations, whereas I want to count each country only once.
How do I summarize this data to have the total number of countries for each members-level?
Here's a sample of my data for reproducibility:
data <-
structure(list(country = structure(c(1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L,
3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 5L,
5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L,
5L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L,
6L, 6L, 6L), .Label = c("A", "B", "C", "D", "E", "F"), class = "factor"),
year = c(1989L, 1990L, 1991L, 1992L, 1993L, 1994L, 1995L,
1996L, 1997L, 1998L, 1999L, 2000L, 2001L, 2002L, 2003L, 2004L,
2005L, 2006L, 2007L, 2008L, 2010L, 1989L, 1990L, 1991L, 1992L,
1993L, 1994L, 1995L, 1996L, 1997L, 1998L, 1999L, 2000L, 2001L,
2002L, 2003L, 2004L, 2005L, 2006L, 2007L, 2008L, 2009L, 2010L,
2011L, 1989L, 1991L, 1993L, 1994L, 1995L, 1996L, 1997L, 1999L,
2000L, 2001L, 2002L, 2003L, 2004L, 2005L, 2006L, 2007L, 2008L,
2010L, 1989L, 1990L, 1991L, 1992L, 1993L, 1994L, 1995L, 1996L,
1997L, 1998L, 1999L, 2000L, 2001L, 2002L, 2003L, 2004L, 2005L,
2006L, 2007L, 2008L, 2009L, 2010L, 2011L, 1991L, 1992L, 1993L,
1994L, 1995L, 1997L, 1998L, 1999L, 2000L, 2001L, 2002L, 2003L,
2004L, 2005L, 2006L, 2007L, 2008L, 2010L, 1991L, 1992L, 1993L,
1994L, 1995L, 1997L, 1998L, 1999L, 2000L, 2001L, 2002L, 2003L,
2004L, 2005L, 2006L, 2007L, 2008L, 2010L), score = c(0L,
0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L,
0L, 0L, 0L, 0L, 0L, 1L, 0L, 1L, 1L, 0L, 1L, 1L, 0L, 1L, 1L,
1L, 1L, 1L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 1L, 2L, 2L,
2L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L,
2L, 0L, 1L, 1L, 1L, 0L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L,
1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 2L, 2L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L
), members = c(7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L,
7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 7L, 6L, 6L, 6L, 6L, 6L,
6L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 7L, 7L, 7L, 7L, 7L, 7L, 7L,
7L, 7L, 7L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 7L, 7L, 7L, 7L, 7L,
7L, 7L, 7L, 7L, 7L, 7L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L,
4L, 4L, 4L, 4L, 5L, 5L, 5L, 6L, 6L, 6L, 6L, 6L, 6L, 6L, 3L,
3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 4L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L)), .Names = c("country", "year", "score",
"members"), class = "data.frame", row.names = c(NA, -121L))
I believe you need this:
library(reshape2)
dcast(aggregate(country~score+members, data=data, FUN=function(x) length(unique(x))),
score~members, value.var="country", fill=0L)
# score 1 2 3 4 5 6 7
#1 0 0 1 0 0 0 1 2
#2 1 1 1 2 2 1 3 2
#3 2 1 0 1 2 0 1 1
Or, to put it the dplyr/tidyr way:
data %>%
group_by(members, score) %>%
summarise(n=n_distinct(country)) %>%
spread(members, n, fill=0L)
## A tibble: 3 x 8
# score 1 2 3 4 5 6 7
#* <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#1 0 0 1 0 0 0 1 2
#2 1 1 1 2 2 1 3 2
#3 2 1 0 1 2 0 1 1
As the OP is using dplyr methods, we can do this by grouping with 'score', 'members' to get the number of elements (n()), and then spread (from tidyr) to reshape it to 'wide' format.
library(dplyr)
library(tidyr)
data %>%
group_by(score, members) %>%
summarise(n = n()) %>%
mutate(members = paste0("m", members)) %>%
spread(members, n, fill = 0)
# score m1 m2 m3 m4 m5 m6 m7
# <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#1 0 0 2 0 0 0 3 30
#2 1 15 3 11 11 3 18 3
#3 2 3 0 2 2 0 6 9
If we need to also get the counts by 'country', just add 'country' in the group_by
data %>%
group_by(country, score, members) %>%
summarise(n = n()) %>%
mutate(members = paste0("m", members)) %>%
spread(members, n, fill = 0)
If the expected output is the one showed in the other posts, an option using data.table would be to convert the 'data.frame' to 'data.table' (setDT(data), and dcast from 'long' to 'wide' specifying the fun.aggregate as uniqueN of the 'value.var' variable i.e. 'country' where uniqueN returns the length of unique elements in the 'country' column. The fill=0 specifies to occupy 0 for those combinations that are not available. By default, it returns as NA.
library(data.table)
dcast(setDT(data), score~members, value.var= 'country', fun.aggregate = uniqueN, fill = 0)
# score 1 2 3 4 5 6 7
#1: 0 0 1 0 0 0 1 2
#2: 1 1 1 2 2 1 3 2
#3: 2 1 0 1 2 0 1 1
It seems the crux of the issue is having the duplicated rows for each year? In which case you can remove them with distinct, then it's a simple crosstab. You could use the %$% exposition pipe from magrittr:
library(dplyr)
library(magrittr)
data %>%
distinct(country, score, members) %$%
table(score, members)
members
score 1 2 3 4 5 6 7
0 0 1 0 0 0 1 2
1 1 1 2 2 1 3 2
2 1 0 1 2 0 1 1
Or a regular pipe and tabyl from the janitor package:
library(dplyr)
library(janitor)
data %>%
distinct(country, score, members) %>%
tabyl(score, members)
score 1 2 3 4 5 6 7
0 0 1 0 0 0 1 2
1 1 1 2 2 1 3 2
2 1 0 1 2 0 1 1
I have a dataframe like this one:
> dput(df)
structure(list(OBBLIGATORIO = structure(c(2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L), .Label = c("no",
"yes"), class = "factor"), COUNTRY = structure(c(16L, 16L, 16L,
16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L,
16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L,
16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L,
16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L,
16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L,
16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L,
16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L, 16L,
16L, 16L, 16L, 16L), .Label = c("Austria", "Belgium", "Bulgaria",
"Croatia", "Cyprus", "Czech Republic", "Denmark", "Estonia",
"Finland", "France", "Germany", "Greece", "Hungary", "Iceland",
"Ireland", "Italy", "Latvia", "Lithuania", "Luxembourg", "Malta",
"Norway", "Poland", "Portugal", "Romania", "Slovakia", "Slovenia",
"Spain", "Sweden", "United Kingdom of Great Britain and Northern Ireland"
), class = "factor"), YEAR = c(2003L, 2006L, 2007L, 2008L, 2009L,
2010L, 1995L, 1996L, 1997L, 1998L, 1999L, 2000L, 2001L, 2002L,
2003L, 2006L, 2007L, 2008L, 2009L, 2010L, 1995L, 1996L, 1997L,
1998L, 1999L, 2000L, 2001L, 2002L, 2003L, 2006L, 2007L, 2008L,
2009L, 2010L, 1995L, 1996L, 1997L, 1998L, 1999L, 2000L, 2001L,
2002L, 2003L, 2006L, 2007L, 2008L, 2009L, 2010L, 1995L, 1996L,
1997L, 1998L, 1999L, 2000L, 2001L, 2002L, 2003L, 2006L, 2007L,
2008L, 2009L, 2010L, 1995L, 1996L, 1997L, 1998L, 1999L, 2000L,
2001L, 2002L, 2003L, 2006L, 2007L, 2008L, 2009L, 2010L, 1995L,
1996L, 1997L, 1998L, 1999L, 2000L, 2001L, 2002L, 2003L, 2006L,
2007L, 2008L, 2009L, 2010L, 1995L, 1996L, 1997L, 1998L, 1999L,
2000L, 2001L, 2002L), AGE = structure(c(1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L), .Label = "Total", class = "factor"),
`CAUSE OF DEATH` = c("Acute poliomyelitis", "Acute poliomyelitis",
"Acute poliomyelitis", "Acute poliomyelitis", "Acute poliomyelitis",
"Acute poliomyelitis", "Acute poliomyelitis", "Acute poliomyelitis",
"Acute poliomyelitis", "Acute poliomyelitis", "Acute poliomyelitis",
"Acute poliomyelitis", "Acute poliomyelitis", "Acute poliomyelitis",
"Diphtheria", "Diphtheria", "Diphtheria", "Diphtheria", "Diphtheria",
"Diphtheria", "Diphtheria", "Diphtheria", "Diphtheria", "Diphtheria",
"Diphtheria", "Diphtheria", "Diphtheria", "Diphtheria", "Measles",
"Measles", "Measles", "Measles", "Measles", "Measles", "Measles",
"Measles", "Measles", "Measles", "Measles", "Measles", "Measles",
"Measles", "Tetanus", "Tetanus", "Tetanus", "Tetanus", "Tetanus",
"Tetanus", "Tetanus", "Tetanus", "Tetanus", "Tetanus", "Tetanus",
"Tetanus", "Tetanus", "Tetanus", "Tuberculosis", "Tuberculosis",
"Tuberculosis", "Tuberculosis", "Tuberculosis", "Tuberculosis",
"Tuberculosis", "Tuberculosis", "Tuberculosis", "Tuberculosis",
"Tuberculosis", "Tuberculosis", "Tuberculosis", "Tuberculosis",
"Viral hepatitis", "Viral hepatitis", "Viral hepatitis",
"Viral hepatitis", "Viral hepatitis", "Viral hepatitis",
"Viral hepatitis", "Viral hepatitis", "Viral hepatitis",
"Viral hepatitis", "Viral hepatitis", "Viral hepatitis",
"Viral hepatitis", "Viral hepatitis", "Whooping cough", "Whooping cough",
"Whooping cough", "Whooping cough", "Whooping cough", "Whooping cough",
"Whooping cough", "Whooping cough", "Whooping cough", "Whooping cough",
"Whooping cough", "Whooping cough", "Whooping cough", "Whooping cough"
), VALUE = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 4L, 2L, 2L, 2L, 1L, 1L, 6L, 7L, 7L, 1L, 2L,
3L, 2L, 5L, 12L, 9L, 13L, 9L, 13L, 8L, 17L, 14L, 16L, 18L,
15L, 19L, 11L, 10L, 25L, 24L, 21L, 22L, 23L, 20L, 34L, 32L,
31L, 30L, 29L, 28L, 27L, 26L, 41L, 42L, 43L, 45L, 46L, 47L,
33L, 35L, 36L, 37L, 38L, 39L, 40L, 44L, 1L, 2L, 1L, 1L, 1L,
2L, 2L, 2L, 1L, 3L, 1L, 1L, 1L, 1L), .Label = c("0", "1",
"2", "3", "6", "7", "9", "17", "18", "19", "21", "22", "27",
"28", "30", "31", "37", "41", "42", "301", "329", "333",
"344", "350", "396", "413", "415", "460", "517", "558", "597",
"609", "622", "647", "681", "1087", "1349", "1413", "1448",
"1499", "1576", "1654", "1725", "1948", "2531", "2665", "2757"
), class = "factor"), ID = 1:98), .Names = c("OBBLIGATORIO",
"COUNTRY", "YEAR", "AGE", "CAUSE OF DEATH", "VALUE", "ID"), row.names = c(NA,
-98L), class = "data.frame")
I want to obtain a chart that:
on x axis there are values from YEAR column
on y axis there are
values from VALUE column data are divided by CAUSE OF DEATH column
So something like:
I try:
x11()
ggplot(df, aes(x = df$`YEAR`, y = df$`VALUE`, fill = df$`CAUSE OF DEATH`, colour = df$`CAUSE OF DEATH`)) +
geom_density(alpha = 0.1) +
xlim(1995, 2010)
But the result is completely different from the one I want.
Thanks
I'm not sure what your actual question is, but one problem with your dataframe is that the VALUE column is currently defined as a factor, not as as a numeric. I think that remedying this will go a long way to solving your problem. I do this post-facto below (i.e. after the dataframe is already created), but if you are getting the data into R via a read.table() or similar command, you can specify the class of your columns at data frame creation time, which is probably a better approach.
In my code below I use the dplyr package for manipulating dataframes. It's quite powerful, but for this particular example it isn't doing anything that base R couldn't do.
require(ggplot2)
require(dplyr)
require(magrittr)
df <- ### YOUR dput output goes here ###
# fix the problem with the `VALUE` column
df %<>% mutate(VALUE = VALUE %>% as.character %>% as.numeric)
# equivalent in base R:
# df$VALUE <- as.numeric(as.character(df$VALUE))
# make a graph (is it the one you want?)
df %>% group_by(YEAR, `CAUSE OF DEATH`) %>%
summarize(value = sum(VALUE)) %>%
ggplot(aes(x = YEAR, y = value, color = `CAUSE OF DEATH`)) +
geom_line() +
theme_bw() +
geom_point()
# save graph for uploading to SO
ggsave('SO37230266.png')
The result is this graph:
I am making a bar plot using lattice in R where I have data for 4 different years on sources of irrigation for different states. using my code, the bar plot is coming fine but I wish the bar corresponding to the year 1996 to be plotted first followed by the bar corresponding to year 2001 etc. so as to show the increasing area being irrigated by tube-wells. However, I am unable to change the ordering. Here is my data and the R code. Many thanks for your help.
# sample data
irr_atlas <- structure(list(state = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L, 3L, 3L,
3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L), .Label = c("ANDHRA PRADESH",
"KARNATAKA", "MADHYA PRADESH", "RAJASTHAN"), class = "factor"),
st_code = c(28L, 28L, 28L, 28L, 28L, 28L, 28L, 28L, 28L,
28L, 28L, 28L, 28L, 28L, 28L, 28L, 29L, 29L, 29L, 29L, 29L,
29L, 29L, 29L, 29L, 29L, 29L, 29L, 29L, 29L, 29L, 29L, 23L,
23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L, 23L,
23L, 23L, 23L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L,
8L, 8L, 8L, 8L, 8L), year = c(1996L, 1996L, 1996L, 1996L,
2001L, 2001L, 2001L, 2001L, 2006L, 2006L, 2006L, 2006L, 2011L,
2011L, 2011L, 2011L, 1996L, 1996L, 1996L, 1996L, 2001L, 2001L,
2001L, 2001L, 2006L, 2006L, 2006L, 2006L, 2011L, 2011L, 2011L,
2011L, 1996L, 1996L, 1996L, 1996L, 2001L, 2001L, 2001L, 2001L,
2006L, 2006L, 2006L, 2006L, 2011L, 2011L, 2011L, 2011L, 1996L,
1996L, 1996L, 1996L, 2001L, 2001L, 2001L, 2001L, 2006L, 2006L,
2006L, 2006L, 2011L, 2011L, 2011L, 2011L), irr_area = c(1.84066,
0.942819, 0.82886, 0.853502, 1.54922, 0.825659, 0.542492,
1.53412, 1.72969, 0.70271, 0.637221, 1.53894, 1.99893, 0.678425,
0.819829, 1.70708, 0.921594, 0.231669, 0.316999, 0.358529,
0.91339, 0.207157, 0.426549, 0.481061, 0.921255, 0.18192,
0.426145, 0.547193, 0.930802, 0.148065, 0.377149, 1.51843,
1.59425, 0.112145, 2.67683, 0.540054, 1.48056, 0.030502,
1.63696, 0.563948, 1.12595, 0.058667, 2.46494, 1.15004, 1.10444,
0.157069, 2.64378, 2.14177, 1.55814, 0.106623, 2.71347, 0.644683,
1.35746, 0.030586, 2.41845, 0.935234, 1.76933, 0.054374,
2.46197, 1.76918, 1.62587, 0.050299, 2.14737, 2.82708),irr_source = structure(c(1L,2L, 4L, 3L, 1L, 2L, 4L, 3L, 1L, 2L, 4L, 3L, 1L, 2L, 4L, 3L,
1L, 2L, 4L, 3L, 1L, 2L, 4L, 3L, 1L, 2L, 4L, 3L, 1L, 2L, 4L,
3L, 1L, 2L, 4L, 3L, 1L, 2L, 4L, 3L, 1L, 2L, 4L, 3L, 1L, 2L,
4L, 3L, 1L, 2L, 4L, 3L, 1L, 2L, 4L, 3L, 1L, 2L, 4L, 3L, 1L,
2L, 4L, 3L), .Label = c("Canal", "Tank", "Tube", "Well"), class = "factor")), .Names = c("state","st_code", "year", "irr_area", "irr_source"), class = "data.frame", row.names = c(NA, -64L))
Code for plot...
library(lattice)
barchart(~irr_area | factor(state) + factor(irr_source),
group=year, data=irr_atlas, auto.key=list(space="right"))
As mentioned, ordering of groups in R graphics is usually determined by the ordering of the factor variable. So, you can reorder your factors with factor and its levels argument.
library(lattice)
barchart(~irr_area | factor(state) + factor(irr_source),
group=factor(year, levels=sort(unique(year), decreasing=T)), # change the order of years
data=irr_atlas, auto.key=list(space="right"))
You can switch it back the other way by changing decreasing=F.
I'm trying to program something quite simple (I think) in R, but I can't seem to get it right. I have a dataset of 50 countries (1 to 50) for 15 years each and about 20 variables per country. For now I am only testing one variable (OS) on my dependent variable (SMD). I would like to do this with a loop country by country so I would get the output for each country in stead of the overall output.
I thought it would be wise to create a subset first (to be able to look at country 1 first, after which my loop should increase the number for country and test country 2). I believe my regression at the bottom of the page should give me the output for country 1 in stead of the overall score for the entire dataset. However I keep getting these errors:
> pdata <- plm.data(newdata, index=c("Country","Date"))
series are constants and have been removed
> pooling <- plm(Y ~ X, data=pdata, model= "pooling")
series Country, xRegion are constants and have been removed
Error in model.matrix.pFormula(formula, data, rhs = 1, model = model, :
NA in the individual index variable
> summary(pooling)
Error in summary(pooling) : object 'pooling' not found
I might be looking at this all wrong, but I believe that without getting this to work, there is no point in going further with programming the loop itself. Any advice on solving my errors, or other ways of programming a loop are really appreciated.
My code:
rm(list = ls())
mydata <- read.table(file = file.choose(), header = TRUE, dec = ",")
names(mydata)
attach(mydata)
Y <- cbind(SMD)
X <- cbind(OS)
newdata <- subset(mydata, Country %in% c(1))
newdata
pdata <- plm.data(newdata, index=c("Country","Date"))
pooling <- plm(Y ~ X, data=pdata, model= "pooling")
summary(pooling)
Edit: data sample of first 2 countries which causes same error
dput(mydata)
structure(list(Region = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L), .Label = c("NAF", "SAME"), class = "factor"), Country = c(1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L), Date = c(1995L, 1996L, 1997L, 1998L,
1999L, 2000L, 2001L, 2002L, 2003L, 2004L, 2005L, 2006L, 2007L,
2008L, 2009L, 2010L, 2011L, 2012L, 2013L, 2014L, 1995L, 1996L,
1997L, 1998L, 1999L, 2000L, 2001L, 2002L, 2003L, 2004L, 2005L,
2006L, 2007L, 2008L, 2009L, 2010L, 2011L, 2012L, 2013L, 2014L
), OS = structure(c(19L, 25L, 27L, 15L, 22L, 20L, 23L, 9L, 7L,
5L, 2L, 1L, 4L, 3L, 6L, 10L, 11L, 13L, 11L, 8L, 26L, 25L, 31L,
29L, 28L, 21L, 30L, 24L, 24L, 16L, 11L, 14L, 12L, 17L, 18L, 29L,
32L, 32L, 33L, 34L), .Label = c("51.5", "52.2", "55.6", "56.4",
"56.7", "57.7", "57.8", "58.3", "59", "59.2", "59.6", "59.9",
"60.2", "60.4", "61.1", "61.2", "62.2", "62.3", "62.8", "63.2",
"63.3", "63.8", "63.9", "64.2", "64.3", "64.5", "64.7", "65.3",
"65.5", "65.6", "66.4", "68", "69.6", "70.7"), class = "factor"),
SMD = structure(c(7L, 12L, 20L, 21L, 17L, 15L, 13L, 10L,
14L, 22L, 23L, 33L, 1L, 32L, 29L, 34L, 28L, 25L, NA, NA,
9L, 6L, 8L, 4L, 2L, 35L, 3L, 36L, 5L, 11L, 16L, 18L, 24L,
19L, 26L, 31L, 27L, 30L, NA, NA), .Label = c("100.3565662",
"13.44788845", "13.45858747", "13.56815534", "15.05892471",
"17.63789658", "18.04088718", "18.3101351", "19.34226196",
"21.25530884", "21.54423145", "23.75898948", "24.08770926",
"26.39817342", "29.44079001", "31.40605191", "34.46667996",
"34.52913657", "35.66070947", "36.4419931", "39.16875621",
"44.0126137", "45.72949566", "49.13062679", "54.83730247",
"56.87886311", "59.80971583", "60.5658962", "69.20148901",
"70.91362874", "72.64845214", "73.97139238", "75.20140919",
"76.18378138", "9.570435019", "9.867635305"), class = "factor")), .Names = c("Region",
"Country", "Date", "OS", "SMD"), class = "data.frame", row.names = c(NA,
-40L))
Are you sure you need to use plm?? This produces a list of summaries by country.
# convert factors to numeric
mydata$SMD <- as.numeric(mydata$SMD)
mydata$OS <- as.numeric(mydata$OS)
# Using lapply(...)
smry <- lapply(unique(mydata$Country),
function(cntry)
summary(lm(SMD~OS,data=mydata[mydata$Country==cntry,])))
# Same thing, using for loop
smry <- list()
for (cntry in unique(mydata$Country)) {
smry <- list(smry,
summary(lm(SMD~OS,data=mydata[mydata$Country==cntry,])))
}
In your dataset, SMD and OS are factors, which need to be converted to numeric first.