The Signal Decomposition
Information | What's this? |
---|---|
Signal | The target (the observed data) |
Trend | General orientation of the signal. The following types of trend
are indicated: Linear, and Quadratic when the signal rises and falls inconsistantly. Note
In certain cases, a data smoothing technique is applied to the results to help reduce volatility or noise in the data. In this case, the trend and cycle analyses are unified as a trend, with the information given that a data smoothing technique was used. |
Cycles |
Smart Predict can detect fixed length or seasonal
cycles. Fixed length cycles recur every N observation. The
recurrence of seasonal cycles is based on a calendar time unit such
as day, week, month etc. For seasonal cycles, the report shows the
recurrence of the cyclic pattern as well as the time granularity
that makes the cyclic pattern appear. The following seasonal cycles
can be detected:
|
Fluctuation | Part of the signal detected by the predictive model that is completely dependent on past values of the signal up to a certain point in time. The influence of the last observations before the predictive forecast is indicated for the signal. |
Residuals | What is left when the trend, the cycles, and the fluctuation have been extracted from the signal. It results from short term fluctuations in the series, which are neither systematic nor predictable: it's the part of the signal that the predictive model can't explain or model. The smaller the residuals, the better the predictive model. A good predictive model produces residual data that contain no pattern. |
Outliers | An actual signal value is qualified as outlier once its corresponding forecasting error is considered to be abnormal relative to the forecasting error mean observed on the estimation data source. The forecasting error indicator is the absolute difference between the actual and predicted values. This is also called the residue. The residue abnormal threshold is set to 3 times the standard deviation of the residue values on an estimation (or validation) data source. See The Signal Outliers for more information. |