Time Series Plots

Introduction

GAUSS provides functions to make it simple to create time series plots. In this tutorial, we will plot a short section of data from the file tbill_3mo.xlsx. This data file can be found in the GAUSS examples directory, starting in GAUSS version 16. If you do not have the file, you can download the file here.

Load the data

Our first step will be to load a portion of the data from this time series. Make sure that the data file, tbill_3mo.xlsx is in your GAUSS working directory and then load the first two years of data with this command:

//Load 24 monthly observations
tbill = xlsReadM("tbill_3mo.xlsx", "B2:B25");

Now the variable, tbill, should contain 24 separate observations. These observations are monthly and the first observation is from January of 1982.

Plot the time series

The GAUSS function plotTS creates time series plots. It takes the following inputs:

myPlot
Optional input, instance of a plotControl structure, containing desired plot attributes.
dstart
Scalar, the first date of the time series. Either a four-digit year, 1982, or a 4 digit year followed by a two-digit month, 198201.
freq
Scalar, the frequency of the data per year. Valid inputs include: 1 annual data, 4 quarterly data, 12 monthly data
y
NxK matrix containing one or more time series to plot.

Let's draw a basic time series plot of our tbill data from above:

//First data in series is January 1982
//dstart = 198201 would also work
dstart = 1982;

//Data is monthly, or 12 times per year
freq = 12;

//Create basic time series plot
plotTS(dstart, freq, tbill);

After running the above code, you should see a plot that looks similar to this:

Basic GAUSS time series plot.

Customize

Now we will customize this plot. If you are not familiar with the GAUSS plotControl structure, or the GAUSS plotSet functions, read through this tutorial to learn how easy they are to use.

Tick label frequency

Let's start our customization by controlling the frequency and location of the tick labels. The plotSetXTicInterval function will allow us to accomplish this task. plotSetXTicInterval takes the following inputs:

&myPlotA pointer to a `plotControl` structure, containing the settings for your plot.
ticIntervalScalar, the number of time periods between tick labels.
firstLabeledOptional input, the first time period on which a label should be placed.
//Declare 'myPlot' to be a plotControl structure
struct plotControl myPlot;

//Fill 'myPlot' with default values for 'xy' plots
//time series plots use 'xy' settings
myPlot = plotGetDefaults("xy");

//Draw the tick labels every 3 months
//Draw the first label at April of 1982
plotSetXTicInterval(&myPlot, 3, 198204);

//Draw graph using plotControl structure
plotTS(myPlot, 1982, 12, tbill);

This time, our graph should look like this:

GAUSS time series plot with custom tick label placement.

As we expect, the first tick label is placed at the fourth month of 1982 and we see one tick label every three months.

Tick label format

Since our tick labels are placed at the start of quarters, we might want to change the tick label format to show the quarter rather than the month. We can do this with plotSetXTicLabel, which takes the following inputs:

&myPlotA pointer to a `plotControl` structure, containing the settings for your plot.
fmtString, a `dttostr` date format specifier. For example, July of 1982 could be represented as:
"YYYY-MO""1982-07"
"YYYY-QQ""1982-Q3"
"MO/YYYY""07/1982"
"YYYY""1982"
labelAngleOptional input, the angle at which to print the tick label.
//Change the tick label format to show quarters
plotSetXTicLabel(&myPlot, "YYYY-QQ");

//Redraw the time series plot
plotTS(myPlot, 1982, 12, tbill);

After the above code, the updated plot will look like this:

GAUSS time series plot with custom, quarterly,  tick labels.

Put it all together

The following code reproduces all settings in this tutorial and also shows how to: set the range of the Y-axis, add a title, add a label to the Y-axis and turn off the background grid:

new;
//Load 24 monthly observations
tbill = xlsReadM("tbill_3mo.xlsx", "B2:B25");

//First data in series is January 1982
//dstart = 198201 would also work
dstart = 1982;

//Data is monthly, or 12 times per year
freq = 12;

//Declare 'myPlot' to be a plotControl structure
struct plotControl myPlot;

//Fill 'myPlot' with default values for 'xy' plots
//time series plots use 'xy' settings
myPlot = plotGetDefaults("xy");

//Draw the tick labels every 3 months
//Draw the first label at April of 1982
plotSetXTicInterval(&myPlot, 3, 198204);

//Change the tick label format to show quarters
plotSetXTicLabel(&myPlot, "YYYY-QQ");

//Add title
plotSetTitle(&myPlot, "US 3 month tbill rates", "Arial", 16);

//Set Y-label
plotSetYLabel(&myPlot, "Percent");

//Set range of Y-axis from 0 to 16
plotSetYRange(&myPlot, 0, 16);

//Turn off background grid
plotSetGrid(&myPlot, "off");

//Redraw the time series plot
plotTS(myPlot, dstart, freq, tbill);

Which will produce a graph that looks like this: Customize GAUSS time series graph.

Have a Specific Question?

Get a real answer from a real person

Need Support?

Get help from our friendly experts.

Try GAUSS for 14 days for FREE

See what GAUSS can do for your data

© Aptech Systems, Inc. All rights reserved.

Privacy Policy