Understanding Data Visualizations: Candlesticks

There are some really neat new visualizations available on the Oracle Analytics Store (formerly Oracle BI Public Store; new link, as well) that I hope you are following if you are a Data Visualization user. …But it hit me a couple weeks ago that there are some more complex visualizations that an end user may not understand. So, here I am with a new series explaining how different visualizations are used.

The first visualization I am going to walk you through is the Candlestick chart. If you are a pricing or financial analyst (especially with daily changes in prices), this visualization should be on your go-to list for analysis. So, what is a candlestick chart? A candlestick chart excels at visualizing price movements over time. It’s kind of a combination of a line and bar chart that gives you 4 pieces of information: Open Price, Close Price, High, and Low. Although they look a lot like boxplots, they are not the same. Candlesticks are used to visualize price and currency patterns over time.

Let’s look at an example…

I’ve loaded 4 years of Oracle stock data that includes the daily open, close, high, low, and volume. If I use traditional visualizations, I get busy screens and compressed information that doesn’t tell the whole story very well.

The first victim is the standard line chart. Line charts are great for comparing numerical values along a timeline. In the visualization below, I enabled a 2^nd Y axis to allow the visualization to show volume on a larger scale than stock price. This tells us quite a bit, but not details of each day’s activity.

The second victim is the combo chart. While not too different from the line chart, sometimes showing area (or perhaps a bar chart) for one of the data elements tells a strong story. Again, I utilized a 2^nd Y axis for volume. While still telling the full story, it’s compressed and hard to decipher. Now, I could add a filter to pare down to a certain date range, but the visualization is still too busy to make sound decisions.

The third victim is a calendar heatmap. The heat map is great at showing cyclical trends visually. Below, we can see that the close price has “warmed up” or increased over some stale points. However, we cannot compare that to the stock open, close, high, or low for each day. We have no clue as to certain day or days of volatility or if that volatility is cyclical, or monthly, quarterly, or yearly trends. Again, not the whole picture.

Finally, we have the candlestick chart. The first thing we see is how clean the canvas is compared to the previous charts. Let’s dig into the pieces of the candlestick chart.

Let’s break down the following slice of the screen:

Here’s where the candlestick label originates. The light gray bar represents the high and low prices of the day, showing the range of prices for that day. The top orange or dark gray bar represents the open and close prices. If the bar is orange, then the close price decreased from the previous day; dark gray means the close price increased. Clearly, some days showed more volatility than others.
The second chart in the visualization shows the volume for the day directly below the candlestick. This is great for determining volume cycles over time if there are any.
Shows the high price along with a calendar.

Note a couple visualization pieces…

See the box in the lower right-hand corner? This is a sliding bar that you can use to increase or decrease the days shown on the canvas.