Research Highlights - February 2020

Comments Off on Research Highlights - February 2020
Research Highlights - February 2020 LoadingADD TO FAVORITES

How can you use the findings from the latest research studies to improve your performance and the performance of your organization?  We’ll provide the highlights and key ideas you won’t get anywhere else.

Researchers at the University of Tokyo and Kozo Keikaku Engineering Inc. have developed a method for enhancing the power of existing algorithms to forecast the future of unknown time series.  By combining the predictions of many suboptimal forecasts, they have been able to construct a consensus prediction that outperforms existing methods.  This research may help provide early warnings for floods, economic shocks, or changes in the weather.

Time series data are a familiar part of our daily lives. A gyrating graph might represent the water level of a river, the price of a stock, or the daily high temperature in a city, just to name a few. Advance knowledge of the future movements of a time series could be used to avert or prepare for future undesirable events.  However, forecasting is extremely difficult because the underlying dynamics that generate the values are nonlinear, even if assumed to be deterministic, and therefore subject to wild fluctuations.

Delay embedding is a widely used method to help make sense of time series data and attempt to predict future values.  This approach takes a sequence of observations and “embeds” them in a higher-dimensional space by combining the current value with evenly spaced lagged values from the past.  For example, to create a three-dimensional delay embedding of the S&P 500 closing price, you can take the closing prices today, yesterday, and the day before as the x-, y-, and z-coordinates, respectively.  However, the possible choices for embedding dimension and delay lags make finding the most useful representation for making forecasts a matter of trial and error.

The Japanese researchers devised a way to select and optimize a collection of delay embeddings so that their combined forecast does better than any individual predictor.  They found that the ‘wisdom of the crowd,’ in which the consensus prediction is better than each on its own, can be true even with mathematical models.  The researchers tested their method on real-world flood...

To continue reading, become a paid subscriber for full access.
Already a Business Briefings subscriber? Login for full access now.

Subscribe for as low as $135/year

  • Get 12 months of Business Briefings that will impact your business and your life
  • Gain access to the entire Business Briefings Research Library
  • Optional Business Briefings monthly CDs in addition to your On-Line access
  • If you do not like what you see, you can cancel anytime and receive a 100% pro-rata refund