[R] ggplot barplot error

[Dennis Murphy]

Hi:

There are several options, but you shot yourself in the foot by using
cbind() to combine objects of different classes into a matrix, so
everything ends up being a factor because a matrix is an atomic object
with a dim attribute, and atomic objects must have a common class, so
Flows and Category are coerced to factors. What you want is to combine
your atomic objects into a data frame, since that is what ggplot()
requires:

#------
Date <- c("2014-04-01","2014-04-02","2014-04-03","2014-04-04","2014-04-07",
          "2014-04-09","2014-04-10","2014-04-01","2014-04-02","2014-04-03",
          "2014-04-04","2014-04-07","2014-04-09","2014-04-10")

Flows <- c(479.6, 187.2,  148.6,   41.5,  123.5,  176.3,   68.3,  401.5,
           -164.2, -195.5, 35.1, -224.0,  -58.6, -138.9)

Category <- c("Equity", "Equity", "Equity", "Equity", "Equity", "Equity",
              "Equity", "Debt", "Debt",   "Debt",   "Debt",   "Debt",
              "Debt",  "Debt")

# Use data.frame, not cbind
DF <- data.frame(Date, Flows, Category)

#-----

Your ggplot() code will "work", but probably not the way you intended.
Here are a couple of versions of your initial plot:

#------
library(ggplot2)

# Initial plot, stacked
ggplot(data = DF, aes(x = Date, y = Flows, fill = Category)) +
    geom_bar(stat="identity")

# Initial plot, dodged
ggplot(data = DF, aes(x = Date, y = Flows, fill = Category)) +
    geom_bar(stat="identity", position = "dodge")
#------


I'm guessing what you want is to 'stack' the dodged version above; if
so, you need to plot the two groups separately and create a fill
factor 'on the fly':

#---
# "Stacking" the dodged plot - requires separate calls
# to each subset so that zero is the anchor

ggplot(data = DF, mapping = aes(x = Date, y = Flows)) +
    geom_bar(data = subset(DF, Category == "Equity"),
             stat = "identity", aes(fill = "Equity")) +
    geom_bar(data = subset(DF, Category == "Debt"),
             stat = "identity", aes(fill = "Debt")) +
    scale_fill_manual("Category",
                      values = c("Equity" = "blue", "Debt" = "red"))
#---


If you want to plot on the actual dates rather than the given dates,
and perhaps to change the date format since the year is common, here
is one way to do it:

#---
# Convert to Date format so that we can plot by actual date
DF$Date <- as.Date(as.character(DF$Date))

# Need scales package to change date format
library(scales)

ggplot(data = DF, mapping = aes(x = Date, y = Flows)) +
    geom_bar(data = subset(DF, Category == "Equity"),
             stat = "identity", aes(fill = "Equity")) +
    geom_bar(data = subset(DF, Category == "Debt"),
             stat = "identity", aes(fill = "Debt")) +
    scale_fill_manual("Category",
                      values = c("Equity" = "blue", "Debt" = "red")) +
    scale_x_date(labels = date_format("%m-%d"))
#---



I hope this covers your use case.

Dennis

On Thu, May 15, 2014 at 4:32 AM, Vikram Bahure
<economics.vikram at gmail.com> wrote:
> Hi,
>
> I am facing issue with ggplot plotting. I have given the data and the code
> below. I have also attached the graph below.
>
> Problem:
> On date "*2014-04-07*", we have debt and equity flow, but it does not shows
> equity flow. Ideally what I would like is the stack for debt/equity starts
> from X axis.
>
> *## Data*
> *Date <-
> c("2014-04-01","2014-04-02","2014-04-03","2014-04-04","2014-04-07",*
> *
> "2014-04-09","2014-04-10","2014-04-01","2014-04-02","2014-04-03",*
> *          "2014-04-04","2014-04-07","2014-04-09","2014-04-10")*
> *Flows <- c(479.6,187.2,  148.6,   41.5,  123.5,  176.3,   68.3,  401.5,*
> *           -164.2, -195.5, 35.1, -224.0,  -58.6, -138.9)*
> *Category <- c("Equity", "Equity", "Equity", "Equity", "Equity", "Equity",*
> *              "Equity", "Debt", "Debt",   "Debt",   "Debt",   "Debt",*
> *              "Debt",  "Debt")*
> *data <- cbind(Date,Flows,Category)*
>
> *## GGplot*
> *ggplot(data = fii.df, aes(x = Date, y = FII, fill = Category)) +
> geom_bar(stat="identity") *
>
> Thanks in advance.
>
> Regards
> Vikram Bahure
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.