Comparative Study

This master thesis explores various comparison methods for multivariate time series in the application area of stock markets. The data is usually compared by juxtaposition or by superimposition. But even a large enough difference between the price ranges can make a simple comparison of the data difficult.

Stock market data is not limited to stocks. It also includes stock (market) indices such as the Dow Jones and NASDAQ. Stock indices have their own proprietary unit, which is different to the unit of a stock. Time series which have different units are also called heterogeneous data. To compare heterogeneous data by superimposition multiple y-axes are often used. But in most cases the arrangement of the axes makes comparisons between different variables completely arbitrary and comparisons are often misleading. The visualization pioneer Jacques Bertin has studied this problem. One suggestion is to index the values, which transforms all data into values which reflect the percent change compared to an indexing point. The first part of the research is concerned with the identification of relevant comparison methods for multivariate time series. The major part of the research deals with the evaluation of an advanced comparison method based on indexing. Another important part is the investigation of the importance of the used axis scale. Differences between linear and logarithmic scale are analyzed for effects on user performance. The major part of this work is a comparative study about three visual comparison methods (visualization types) for multivariate time series. The three tested comparison methods are juxtaposition, superimposition and indexing. 24 test persons participated in the study. Each participant had to complete 14 tasks for each one of the three visualization types. The task completion time and the task correctness for every task were measured and later used for statistical analyses. This work further presents state of the art research about other visualizations suited for visual comparison tasks. The prototype application incorporates several common stock market visualizations such as line charts, OHLC charts and candlestick charts. Basic comparison methods like juxtaposition and superimposition are available. A more advanced comparison method based on indexing was implemented. The usability test results support the assumption that the indexing method enables the user to perform comparison tasks with much less estimation errors. The task completion time is not significantly different. The free selection of the indexing point makes comparisons for a certain time period more effective and delivers more precise results. A post-test survey showed that the majority of the participants favor the indexing method over the two other visualization types. The test results for the usage of different scales indicate that tasks were faster completed when using logarithmic scales. The task correctness rate was not significantly different between linear and logarithmic scales.