Automatic trend estimation
Our book introduces a method to evaluate the accuracy of trend estimation algorithms under conditions similar to those encountered in real time series processing. This method is based on Monte Carlo experiments with artificial time series numerically generated by an original algorithm. The second pa...
Clasificación: | Libro Electrónico |
---|---|
Autores principales: | , |
Autor Corporativo: | |
Formato: | Electrónico eBook |
Idioma: | Inglés |
Publicado: |
Dordrecht :
Springer Netherlands : Imprint: Springer,
2013.
|
Edición: | 1st ed. 2013. |
Colección: | SpringerBriefs in Physics,
|
Temas: | |
Acceso en línea: | Texto Completo |
Tabla de Contenidos:
- Discrete stochastic processes and time series
- Trend definition
- Finite AR(1) stochastic process
- Monte Carlo experiments. - Monte Carlo statistical ensembles
- Numerical generation of trends
- Numerical generation of noisy time series
- Statistical hypothesis testing
- Testing the i.i.d. property
- Polynomial fitting
- Linear regression
- Polynomial fitting
- Polynomial fitting of artificial time series
- An astrophysical example
- Noise smoothing
- Moving average
- Repeated moving average (RMA)
- Smoothing of artificial time series
- A financial example
- Automatic estimation of monotonic trends
- Average conditional displacement (ACD) algorithm
- Artificial time series with monotonic trends
- Automatic ACD algorithm
- Evaluation of the ACD algorithm
- A paleoclimatological example
- Statistical significance of the ACD trend
- Time series partitioning
- Partitioning of trends into monotonic segments
- Partitioning of noisy signals into monotonic segments
- Partitioning of a real time series
- Estimation of the ratio between the trend and noise
- Automatic estimation of arbitrary trends
- Automatic RMA (AutRMA)
- Monotonic segments of the AutRMA trend
- Partitioning of a financial time series.