R Dygraphs X-axis granularity from monthly to yearly

Question

I'm using dygraphs and would like to change the x axis granularity from daily to yearly. I have a few daily data points all in the format c("2009-01-01", "2010-01-01", "2011-01-01"), and the x axis ticks appears at Jan 2009, June 2009, Jan 2010, June 2010 I want the x axis to only appear as "2009, 2010, 2011" corresponding to data points that actually exist..

I have the following code:

dygraph(hhinfl) %>% dyLegend(width=400, show="auto") %>% dyOptions(drawPoints=TRUE, pointSize=3) %>% dySeries("Household Inflation", color="#0174DF") %>% dySeries("date", color="#FFFFFF")%>% dySeries("Average Inflation", color="#DF0101") %>% dyAxis("y", label="Inflation Rate (%)")  %>% dyAxis("x", drawGrid = FALSE, axisLabelFormatter="function(d) { return d.getFullYear() }")

which returns the YEAR of each date on the x-axis, but that means there are multiple years like "Jan 2009, June 2009, Jan 2010, June 2010" becomes "2009, 2009, 2010, 2010"

Alternatively, another code shows:

dygraph(hhinfl) %>% dyLegend(width=400, show="auto") %>% dyOptions(drawPoints=TRUE, pointSize=3) %>% dySeries("Household Inflation", color="#0174DF") %>% dySeries("date", color="#FFFFFF")%>% dySeries("Average Inflation", color="#DF0101") %>% dyAxis("y", label="Inflation Rate (%)")  %>% dyAxis("x", drawGrid = FALSE, ticker= " function (a, b, pixels, opts, dygraph, vals) {return Dygraph.getDateAxis(a, b, Dygraph.YEARLY, opts, dygraph);}", axisLabelFormatter="function(d) { return d.getFullYear() }")

Does not return any graph at all, except for the y-axis.

How may I resolve this issue?

hhinfl is an xts file constructed as follows:

dateseq<- seq(from=as.Date("2010-01-01"),to=as.Date("2013-12-31"),by="year")
household<- c(2.3, 2.4, 2.5, 3.1)
avg<- c(2.5, 2.6, 2.7, 3.1)

hhinfl<- data.frame(date=dateseq, hh=household, average=avg)
colnames(hhinfl)<- c("date", "Household Inflation", "Average Inflation")
hhinfl<-xts(hhinfl, order.by=hhinfl$date)

Please provide your data frame or a small part of it (use dput() to produce it). That way answerers can run code to test their answers. — lawyeR, Nov 24 '15 at 13:31
@lawyeR thank you for the comments, I have now included the necessary code to construct the data frame — stochastiq, Nov 25 '15 at 04:35
Error in dySeries(., "Household Inflation", color = "#0174DF") : One or more of the specified series were not found. Valid series names are: date, hh, average. That is the error I get when I run your MWE code. — lawyeR, Nov 25 '15 at 11:07
@lawyeR sorry about that, I have included a line of code for colnames(hhinfl) that should fix it. Let me know if there are any more issues — stochastiq, Nov 26 '15 at 01:15

score 4 · Answer 1 · answered Sep 23 '16 at 15:48

So close! Before I noticed your "Alternatively, another code shows:" above, I wrote code essentially identical to yours (cribbing from http://jsfiddle.net/kaliatech/P8ehg/) and got the same blanked-out chart.

Then I stumbled into https://github.com/danvk/dygraphs/blob/master/src/dygraph-tickers.js#L221 and saw that the parameter should be ANNUAL, not YEARLY.

So here's what turns out well for me:

dygraph(hhinfl) %>% 
    dyLegend(width=400, show="auto") %>% 
    dyOptions(drawPoints=TRUE, pointSize=3) %>% 
    dySeries("Household Inflation", color="#0174DF") %>% 
    dySeries("date", color="#FFFFFF") %>% 
    dySeries("Average Inflation", color="#DF0101") %>% 
    dyAxis("y", label="Inflation Rate (%)")  %>% 
    dyAxis("x", drawGrid = FALSE) %>%
    dyAxis("x", axisLabelFormatter="function(d) { return d.getFullYear() }") %>%
    dyAxis("x", ticker="function(a, b, pixels, opts, dygraph, vals) {
                            return Dygraph.getDateAxis(a, b, Dygraph.ANNUAL, opts, dygraph)
                            // or TWO_HOURLY or SIX_HOURLY...
                        }")

R Dygraphs X-axis granularity from monthly to yearly

1 Answers1