Prediction of trends based on history values of an observable

1. A method comprising:

fitting, by a processor, a mathematical model on history values of an observable, the mathematical model having a ratio factor defined as a ratio between a number of history values used for calculation of a given observable value and a number of observable values modelled simultaneously;

receiving information defining a minimum forecast horizon and a maximum forecast horizon and information defining a minimum ratio factor value and a maximum ratio factor value;

receiving information defining a desired forecast horizon;

determining a desired ratio factor value corresponding to the desired forecast horizon based at least on the desired forecast horizon, the minimum forecast horizon, the maximum forecast horizon, the minimum ratio factor value, and the maximum ratio factor value; and

using the desired ratio factor value in fitting the mathematical model on the history values of the observable.

2. The method of claim 1 , wherein a first desired ratio factor value corresponding to the minimum forecast horizon is smaller than a second desired ratio factor value corresponding to the maximum forecast horizon.

3. The method of claim 1 , wherein the desired ratio factor value is determined by interpolation, wherein the interpolation is based on the desired forecast horizon, the minimum forecast horizon, the maximum forecast horizon, the minimum ratio factor value, and the maximum ratio factor value.

4. The method of claim 1 , wherein the mathematical model is one of the following: autoregressive model, univariate autoregressive model, autoregression with moving average model, autoregression with integrated moving average, and neural networks.

5. The method of claim 1 , wherein the number of history values includes all available history values.

6. The method of claim 1 , wherein the number of history values comprises a set of most recent history values.

7. The method of claim 1 , further comprising periodically updating the history values.

8. The method of claim 1 , further comprising:

determining whether the desired forecast horizon exceeds the maximum forecast horizon; and

in response to determining that the desired forecast horizon exceeds the maximum forecast horizon, dynamically changing a step size in the mathematical model, wherein the change in step size in the mathematical model corresponds to a time difference.

9. The method of claim 8 , wherein dynamically changing the step size includes fitting the mathematical model on a first set of history values and omitting a second set of history values from the mathematical model.

10. The method of claim 1 , further comprising:

scaling the history values from a first step size to a second step size to produce scaled history values, wherein the first step size and the second step size are associated with the mathematical model;

using the desired ratio factor value in fitting the mathematical model on the scaled history values of the observable to produce forecast data; and

scaling the forecast data to the first step size.

11. A data processing system comprising:

a processor;

a future prediction module, executable by the processor, for fitting a mathematical model on history values of an observable, the mathematical model having a ratio factor defined as a ratio between a number of history values used for calculation of a given observable value and a number of observable values modelled simultaneously;

a storage module, executable by the processor, for storing information defining a minimum forecast horizon and a maximum forecast horizon and for storing information defining a minimum ratio factor value and a maximum ratio factor value;

an input module, executable by the processor, for receiving information defining a desired forecast horizon; and

a ratio factor module, executable by the processor, for determining a desired ratio factor value corresponding to the desired forecast horizon based at least on the desired forecast horizon, the minimum forecast horizon, the maximum forecast horizon, the minimum ratio factor value, and the maximum ratio factor value,

wherein the future prediction module is arranged to use the desired ratio factor value in fitting the mathematical model on the history values of the observable.

12. The data processing system of claim 11 , wherein the observable is a key performance indicator of a database management system.

13. A data processing system comprising:

means for fitting a mathematical model on history values of an observable, the mathematical model having a ratio factor defined as a ratio between a number of history values used for calculation of a given observable value and a number of observable values modelled simultaneously;

means for storing information defining a minimum forecast horizon and a maximum forecast horizon and for storing information defining a minimum ratio factor value and a maximum ratio factor value;

means for receiving information defining a desired forecast horizon; and

means for determining a desired ratio factor value corresponding to the desired forecast horizon based at least on the desired forecast horizon, the minimum forecast horizon, the maximum forecast horizon, the minimum ratio factor value, and the maximum ratio factor value,

wherein the means for fitting the mathematical model is arranged to use the desired ratio factor value in the mathematical model, and wherein the means for determining the desired ratio factor value comprises at least one of circuitry configured to determine the desired ratio factor value and a processor configured to execute processor-executable instructions to determine the desired ratio factor value.

14. The data processing system of claim 13 , wherein the observable is a key performance indicator of a database management system.

15. A computer program product stored on a non-transitory computer-usable storage medium having a computer-readable program, where the computer-readable program, when executed on a data processing system, causes the data processing system to:

fit a mathematical model on history values of an observable, the mathematical model having a ratio factor defined as a ratio between a number of history values used for calculation of a given observable value and a number of observable values modelled simultaneously;

receive information defining a minimum forecast horizon and a maximum forecast horizon and information defining a minimum ratio factor value and a maximum ratio factor value;

receive information defining a desired forecast horizon;

determine a desired ratio factor value corresponding to the desired forecast horizon based at least on the desired forecast horizon, the minimum forecast horizon, the maximum forecast horizon, the minimum ratio factor value, and the maximum ratio factor value; and

use the desired ratio factor value in fitting the mathematical model on the history values of the observable.

16. The computer program product of claim 15 , wherein a first desired ratio factor value corresponding to the minimum forecast horizon is smaller than a second desired ratio factor value corresponding to the maximum forecast horizon.

17. The computer program product of claim 16 , wherein the mathematical model is one of the following: autoregressive model, univariate autoregressive model, autoregression with moving average model, autoregression with integrated moving average, and neural networks.

18. The computer program product of claim 15 , wherein the desired ratio factor value is determined using interpolation.

19. The computer program product of claim 18 , wherein the interpolation is based on the desired forecast horizon, the minimum forecast horizon, the maximum forecast horizon, the minimum ratio factor value, and the maximum ratio factor value.

20. The computer program product of claim 15 , wherein the mathematical model is one of the following: autoregressive model, univariate autoregressive model, autoregression with moving average model, autoregression with integrated moving average, and neural networks.