I am using the following code to look at the past 9 months of a stock.
library(quantmod)
getSymbols("AMZN")
candleChart(to.weekly(AMZN),multi.col=TRUE,theme="white",subset='last 9 months')
addADX()
You can see that the red line is essentially not included in the plot because it mostly lies below the value of 20. I want to modify the Y axis range of addADX so that it always shows all three lines. How would it be possible?
The input parameters of addADX() only control the computation of the directional movement index - not the graphical parameters which are set according to the average direction index.
A simple workaround to display the positive and negative direction index completely is to compute the directional movement index by yourself with ADX() from the TTR package and then add it to the previous chart with addTA(), which allows more customisation.
library(quantmod)
getSymbols("AMZN")
dat <- to.weekly(AMZN)
candleChart(dat, multi.col = TRUE, theme = "white", subset = "last 9 months")
adx <- ADX(HLC(dat), n = 14, maType = "EMA", wilder = TRUE)[, c("DIp", "DIn", "ADX")]
addTA(adx, col = c("green", "red", "blue"), lwd = c(1, 1, 2), legend = NULL)
Related
I am a beginner in R and thanks to the provided help here so far, I managed to create the following Barplot by using the code below. However, one issue is still open and I couldn't resolve it so far, even after reading all the answers provided in similar questions.
Code:
res <- data.frame( aggregate( rentals ~ weatherCond + season, bikes, sum ))
res$weatherCond <- ordered(res$weatherCond,
levels = c(1, 2, 3, 4),
labels = c("Clear", "Fog", "Snow", "Storm"))
res$season <- factor(res$season,
levels = c(1, 2, 3, 4),
labels = c("Winter", "Spring", "Summer", "Autumn"))
par(mar=c(10,4,2,1)+.5)
barplot(res[,"rentals"], names=apply( res[,1:2], 1, function(x) paste0(x[1],"_",x[2]) ), las=3, main = "Weather and Rentals", ylab = "Rentals", col = c('blue', 'red', 'green', 'grey')[res$weatherCond])
The plot visualizes the connection between the number of RENTED bikes and the SEASONs depending on the WEATHER conditions (3 variables). The bars in the plot show the amount of bikes RENTED, separated into the corresponding SEASON/WEATHER combinations.
My question:
how exactly should I modify the code above so that the count of the value of every single bar can be displayed (in the bar or on top of the bar)? Every single bar represents the number of rentals.
UPDATE:
I managed to display the values by adding the following lines of code:
y <- as.matrix(res$rentals)
text(b, y+2, labels=as.character(y)) <!-- the "b" is the barplot -->
However, the result is horrifying. Is there a way to make it look better?
barplot() returns the midpoints of each bar (see ?barplot). This can be used in combination with the function text() to display the numbers.
I your case, it would become:
mid_points <- barplot(res[,"rentals"], ...)
text(x = mid_points, y = res[,"rentals"], text = res[,"rentals"])
Further parameters can be used the modify the placement of the numbers.
Since I don't have your data, I used a modified base R example:
b <- barplot(GNP ~ Year, data = longley, ylim = c(0, 700))
text(
x = b,
y = longley$GNP,
labels = round(longley$GNP, 1),
pos = 3,
srt = 90,
offset = 1.5
)
I am currently working with bfastSpatial, I am attempting to plot breakpoint values as a year colour based legend. I am aware of changeMonth function for plotting monthly based breakpoints (http://www.loicdutrieux.net/bfastSpatial/) however, I am attempting to achieve a similar outcome as in Morrison et al. (2019) https://www.mdpi.com/2072-4292/10/7/1075
Any assistance would be appreciated.
If you want to round the breakpoint dates to integer years, you can use floor, as the dates are in decimal years. Next, to make a similar plot as the one you showed, you can use the tmap package. Since you did not attach any data to the OP, I used the tura data included in the bfastSpatial package.
library(bfastSpatial)
library(tmap)
# Load tura data
data(tura)
# Perform bfast analysis
bfm <- bfmSpatial(tura, start=c(2009, 1), order=1)
# Extract the first band (breakpoints)
change <- bfm[[1]]
# As breakpoints dates are in year decimals,
# you can use floor to round them to lowest integer
change <- floor(change)
# Set shape as change, the object to plot
tm_shape(change) +
# Plot it as raster and set the palette, number of categories,
# style (categorical) and title of the legend.
tm_raster(palette = "Spectral",
n = 5,
style = "cat",
title = "Year") +
# Set the legend's position and eliminate the comma used by default for
# separating thousands values. Add background color and transparency
tm_layout(legend.position = c("right", "bottom"),
legend.format=list(fun=function(x) formatC(x, digits=0, format="d")),
legend.bg.color = "white",
legend.bg.alpha = 0.7) +
# Add scale bar, set position and other arguments
tm_scale_bar(breaks = c(0,0.5,1),
position = c("right", "top"),
bg.color = "white",
bg.alpha = 0.7) +
# Add north arrow with additional parameters
tm_compass(type = "arrow",
position = c("left", "top"),
bg.color = "white",
bg.alpha = 0.7)
The obtained plot:
I'm trying to fit Variance-Gamma distribution to empirical data of 1-minute logarithmic returns. In order to visualize the results I plotted together 2 histograms: empirical and theoretical.
(a is the vector of empirical data)
SP_hist <- hist(a,
col = "lightblue",
freq = FALSE,
breaks = seq(a, max(a), length.out = 141),
border = "white",
main = "",
xlab = "Value",
xlim = c(-0.001, 0.001))
hist(VG_sim_rescaled,
freq = FALSE,
breaks = seq(min(VG_sim_rescaled), max(VG_sim_rescaled), length.out = 141),
xlab = "Value",
main = "",
col = "orange",
add = TRUE)
(empirical histogram-blue, theoretical histogram-orange)
However, after having plotted 2 histograms together, I started wondering about 2 things:
In both histograms I stated, that freq = FALSE. Therefore, the y-axis should be in range (0, 1). In the actual picture values on the y-axis exceed 3,000. How could it happen? How to solve it?
I need to change the bucketing size (the width of the buckets) and the density per unit length of the x-axis. How is it possible to do these tasks?
Thank you for your help.
freq=FALSE means that the area of the entire histogram is normalized to one. As your x-axis has a very small range (about 10^(-4)), the y-values must be quite large to achieve an area (= x times y) of one.
The only way to set the number of bins is by providing a vector of break points to the parameter breaks. Theoretically, this parameter also accepts a single number, but this number is ignored by hist. Thus try the following:
bins <- 6 # number of cells
breaks <- seq(min(x),max(x),(max(x)-min(x))/bins)
hist(x, freq=FALSE, breaks=breaks)
I was wondering if it's possible to get a two sided barplot (e.g. Two sided bar plot ordered by date) that shows above Data A and below Data B of each X-Value.
Data A would be for example the age of a person and Data B the size of the same person. The problem with this and the main difference to the examples above: A and B have obviously totally different units/ylims.
Example:
X = c("Anna","Manuel","Laura","Jeanne") # Name of the Person
A = c(12,18,22,10) # Age in years
B = c(112,186,165,120) # Size in cm
Any ideas how to solve this? I don't mind a horizontal or a vertical solution.
Thank you very much!
Here's code that gets you a solid draft of what I think you want using barplot from base R. I'm just making one series negative for the plotting, then manually setting the labels in axis to reference the original (positive) values. You have to make a choice about how to scale the two series so the comparison is still informative. I did that here by dividing height in cm by 10, which produces a range similar to the range for years.
# plot the first series, but manually set the range of the y-axis to set up the
# plotting of the other series. Set axes = FALSE so you can get the y-axis
# with labels you want in a later step.
barplot(A, ylim = c(-25, 25), axes = FALSE)
# plot the second series, making whatever transformations you need as you go. Use
# add = TRUE to add it to the first plot; use names.arg to get X as labels; and
# repeat axes = FALSE so you don't get an axis here, either.
barplot(-B/10, add = TRUE, names.arg = X, axes = FALSE)
# add a line for the x-axis if you want one
abline(h = 0)
# now add a y-axis with labels that makes sense. I set lwd = 0 so you just
# get the labels, no line.
axis(2, lwd = 0, tick = FALSE, at = seq(-20,20,5),
labels = c(rev(seq(0,200,50)), seq(5,20,5)), las = 2)
# now add y-axis labels
mtext("age (years)", 2, line = 3, at = 12.5)
mtext("height (cm)", 2, line = 3, at = -12.5)
Result with par(mai = c(0.5, 1, 0.25, 0.25)):
I'm trying to add color to specific points in my circular data based on group membership (I have two groups: one with individuals with a certain medical condition and another group of just healthy controls). I've converted their data from degrees to radians and put it on the plot, but I haven't managed to be able to selectively change the color of the points based on the factor variable I have).
Know that I've loaded library (circular), which doesn't allow me to use ggplot. Here's the syntax I've been working with:
plot(bcirc, stack=FALSE, bins=60, shrink= 1, col=w$dx, axes=FALSE, xlab ="Basal sCORT", ylab = "Basal sAA")
If you've noticed, I specified the factor variable (which has two levels) in the color section, but it just keeps putting everything in one color. Any suggestions?
Seems plot.circular does not like to assign multiple colours. Here's one potential work-around:
library(circular)
## simulate circular data
bcirc1 <- rvonmises(100, circular(90), 10, control.circular=list(units="degrees"))
bcirc2 <- rvonmises(100, circular(0), 10, control.circular=list(units="degrees"))
bcirc <- c(bcirc1, bcirc2)
dx <- c(rep(1,100),rep(2,100))
## start with blank plot, then add group-specific points
plot(bcirc, stack=FALSE, bins=60, shrink= 1, col=NA,
axes=FALSE, xlab ="Basal sCORT", ylab = "Basal sAA")
points(bcirc[dx==1], col=rgb(1,0,0,0.1), cex=2) # note: a loop would be cleaner if dealing with >2 levels
points(bcirc[dx==2], col=rgb(0,0,1,0.1), cex=2)
Inspired by Paul Regular's example, here is a version using the same data where one condition is plotted stacking inwards and the other is plotted stacking outwards.
library(circular)
## simulate circular data
bcirc1 <- rvonmises(100, circular(90, units = 'degrees'), 10, control.circular=list(units="degrees"))
bcirc2 <- rvonmises(100, circular(0, units = 'degrees'), 10, control.circular=list(units="degrees"))
bcirc <- data.frame(condition = c(
rep(1,length(bcirc1)),
rep(2,length(bcirc2)) ),
angles = c(bcirc1,
bcirc2) )
## start with blank plot, then add group-specific points
dev.new(); par(mai = c(1, 1, 0.1,0.1))
plot(circular(subset(bcirc, condition == 1)$angles, units = 'degrees'), stack=T, bins=60, shrink= 1, col=1,sep = 0.005, tcl.text = -0.073,#text outside
axes=T, xlab ="Basal sCORT", ylab = "Basal sAA")
par(new = T)
plot(circular(subset(bcirc, condition == 2)$angles, units = 'degrees'), stack=T, bins=60, shrink= 1.05, col=2,
sep = -0.005, axes=F)#inner circle, no axes, stacks inwards