Visualizing
Time Series

 

Khairi Reda

redak@iu.edu
@RedKhair

Visual representations of
time series

 

  • The familiar: line charts, scatterplots
  • The not so familiar: stripe, and horizon graphs
  • Interaction techniques
  • Graphical perception of times series
  • Tempo: faceted visualization of time series
 

set A

x y
10 8.04
8 6.95
13 7.58
9 8.81
11 8.33
14 9.96
6 7.24
4 4.26
12 10.84
7 4.82
5 5.68

set B

x y
10 9.14
8 8.14
13 8.74
9 8.77
11 9.26
14 8.1
6 6.13
4 3.1
12 9.13
7 7.26
5 4.74

set C

x y
10 7.46
8 6.77
13 12.74
9 7.11
11 7.81
14 8.84
6 6.08
4 5.39
12 8.15
7 6.42
5 5.73

set D

x y
8 6.58
8 5.76
8 7.71
8 8.84
8 8.47
8 7.04
8 5.25
19 12.5
8 5.56
8 7.91
8 6.89
mean
9.00 | 7.50
9.00 | 7.50
9.00 | 7.50
9.00 | 7.50
variance
11.00 | 4.12
11.00 | 4.12
11.00 | 4.12
11.00 | 4.12
correlation
0.816
0.816
0.816
0.816
regression
Y = .5X + 3
Y = .5X + 3
Y = .5X + 3
Y = .5X + 3
R2
0.66
0.66
0.66
0.66

Eco-Spec Time series

Time Temperature
Environmental variables and flux measurements
air temperature = [ 15.0, 15.5, 15.1, 14.7, 16.3, 18.7, 19.8, 20.3, 19.5, 16.9, 18.1, 16.1, 14.1, 15.5, ... ]
rel. humidity   = [ 60.1, 67.1, 62.0, 54.0, 51.8, 61.3, 64.9, 65.5, 63.0, 60.1, 57.2, 51.0, 55.0, 54.1, ... ]
carbon flux     = [ 1.11, 1.22, 1.14, 1.15, 1.44, 1.21, 1.31, 1.12, 1.11, 1.12, 1.15, 1.12, 1.14, 1.12, ... ]
     ...
Hyperspectal optical indices
NDVI          = [ 0.06, 0.05, 0.07, 0.09, 0.07, 0.05, 0.07, 0.09, 0.06, 0.05, 0.05, 0.06, 0.06, 0.05, ... ]
RENDVI        = [ 0.09, 0.10, 0.12, 0.10, 0.11, 0.12, 0.12, 0.11, 0.09, 0.13, 0.09, 0.09, 0.10, 0.09, ... ]
CI            = [ 1.11, 1.22, 1.14, 1.15, 1.44, 1.21, 1.31, 1.12, 1.11, 1.12, 1.15, 1.12, 1.14, 1.12, ... ]
     ...

Analytical tasks

  • Understand association between hyperspectral reflectance and plant physiological processes
  • Analyze the above relationships at different seasonal and diurnal phases
  • Develop models to predict carbon flux based on optical indices

Analyze relationships between multiple time series

Time series
visualization

 

Line charts

William Playfair. Commercial and Political Atlas, 1786

Gross Domestic Product (USD per capita)

Small multiples

Aspect ratio matters

Atmospheric Carbon Dioxide
Average absolute orientation of line segments should be close to 45° for better perception

Multi-scale views

Chicago temperature anomalies

Multi-scale views


Waqas and Elmqvist. Stack Zooming for multi-focus interaction in time-series data visualization

Lenses


Zhao et al. Exploratory Analysis of Time-Series with ChronoLenses. Transactions on Visualization and Computer Graphics 17(12), pp. 2422-2431, 2011

Multi-scale views


Zhao et al. KronoMiner: Using Multi-Foci Navigation for the Visual Exploration of Time-Series Data. CHI'11

Stripe chart

Encode time series value with color as opposed to position

 

 

Dow Jones index







Stripe chart

High frequency noise

 

 


Stripe chart

Electropherograms (112 Wheat strains)

Kincaid and Lam. Line Graph Explorer: Scalable Display of Line Graphs Using Focus+Context, AVI'06

Stripe chart

Electropherograms (112 Wheat strains)

Kincaid and Lam. Line Graph Explorer: Scalable Display of Line Graphs Using Focus+Context, AVI'06

Horizon chart


Temperature anomalies

Data: http://berkeleyearth.lbl.gov/city-list/

Graphical perception of Horizon Charts


Graphical perception of Horizon Charts


Graphical perception of Horizon Charts


Horizon vs. Line chart

  • A 2-band horizon chart is more accurate than a standard line chart
  • Horizon charts take more time to "read" than traditional line charts
  • Increasing the number of bands decreases estimation accuracy
  • The ability to see more time series may outweigh the loss of accuracy

Comparison of multiple time series

  • If the analysis requires comparison across a large visual span, use small-multiples or horizon plots
  • If task requires local comparisons within a short visual span, use simple line or braided charts
  • Small multiples is generally a good choice for intuitiveness and speed

Scatterplots

Scatterplots

Animated scatterplots


Connected scatterplots

Tempo

  • Web-based tool for visualizing time series
  • Show diurnal and seasonal associations between variables
  • Combine scatterplots and line charts
  • Faceted views to limit visual clutter and stratify data temporally

Demo

Future work

  • Automatic classification / ordering of views based on shape-characteristics of scatterplots
  • Provide mechanisms to load larger datasets
  • Improve interactions

Acknowledgments


EVS: Yuki Hamada, Paul Tarpey

ALCF: Mike Papka, Tom Marrinan