LAUSR

We study the application of time series forecasting methods to massive datasets of short financial time series. In our example, the time series arise from analyzing monthly expenses and incomes in personal financial records. Differently from traditional time series forecasting applications, we work with series of very short depth (as short as 24 data points), which prevents from using classical exponential smoothing methods. However, this shortcoming is compensated by the the size of our dataset: millions of time series. The latter allows tackling the problem of time series prediction from a pattern recognition perspective. Specifically, we propose a method for short time series prediction based on time series clustering and distance-based regression. We experimentally show that this strategy leads to improved accuracy compared to exponential smoothing methods. We additionally describe the underlying big data platform developed to carry out the forecasts in an efficient manner (comparisons to millions of elements in near real-time).