Rescaled Range and "Persistence"

  • Thread starter Jimster41
  • Start date
  • Tags
    Range
In summary: Maybe it relates back to similar persistence in rainfall, perhaps connected with the Southern Oscillation Index. Or low rainfall at the source catchment in one season may leave to its being relatively dry at the start of the next wet season, so more water is retained locally and less floods down.
  • #1
Jimster41
783
82
Why is "black noise" 1/(f^B) where B>2 associated with "persistence" and "long range dependence"?
This book I'm reading uses the level of the Nile river as an example... I can understand what the Rescaled Range is saying but i don't get the implications w/respect to those terms or phrases. Is the idea that there is some relaxation process or processes with very long relaxation times? If so doesn't that somehow suggest one Nile flood is connected to another Nile flood by that process? This seems absurd, what process?
 
Physics news on Phys.org
  • #2
Jimster41 said:
Why is "black noise" 1/(f^B) where B>2 associated with "persistence" and "long range dependence"?
This book I'm reading uses the level of the Nile river as an example... I can understand what the Rescaled Range is saying but i don't get the implications w/respect to those terms or phrases. Is the idea that there is some relaxation process or processes with very long relaxation times? If so doesn't that somehow suggest one Nile flood is connected to another Nile flood by that process? This seems absurd, what process?
I can imagine two such processes. Maybe it relates back to similar persistence in rainfall, perhaps connected with the Southern Oscillation Index. Or low rainfall at the source catchment in one season may leave to its being relatively dry at the start of the next wet season, so more water is retained locally and less floods down.
 
  • #3
Thanks, I sort of came to the same conclusion (that there is some causal geophysical system at work over that time scale) last night while trying to figure out why this example of noise, chaos, fractals has me hung up when the dozens of others the author has described have been easier to grok (book is the one by M. Schroeder: "Fractals, Chaos, Power Laws...")

The part that I'm confused about is how you can have a function (the "rescaled range") of delta t (time interval) and still have something that is noise, which I was thinking had to be random in time. In other words (let me phrase this carefully) why doesn't R(deltat), because it is detectably a function of time, imply regular, structured, periodic, wave-like in space-time? Or is it one of those but not all?
 

FAQ: Rescaled Range and "Persistence"

What is rescaled range?

Rescaled range (R/S) is a statistical method used to analyze the long-term trends and patterns in time series data. It was developed by British hydrologist Harold Edwin Hurst in the early 1950s and is commonly used in fields such as economics, finance, and meteorology.

How is rescaled range calculated?

To calculate rescaled range, first the range (R) and standard deviation (S) of a time series data set is calculated. Then, the rescaled range is found by dividing the range by the standard deviation. This value is then plotted against different time scales to determine the presence of long-term trends or patterns.

What is persistence in time series data?

Persistence refers to the degree to which a time series data set displays long-term trends or patterns. If a data set displays high persistence, it means that the values tend to remain above or below the mean for extended periods of time. Low persistence indicates that there is little to no long-term trend or pattern in the data.

How is persistence measured using rescaled range?

Persistence is measured using the Hurst exponent, which is calculated using the rescaled range. A Hurst exponent of 0.5 indicates a lack of persistence, while values above 0.5 indicate high persistence. This measure is used to determine the level of predictability in a time series data set.

What are the limitations of rescaled range analysis?

While rescaled range analysis can provide valuable insights into time series data, it has some limitations. One limitation is that it assumes that the data follows a fractal pattern, which may not always be the case. Additionally, it does not account for external factors that may influence the data, such as economic or environmental events.

Back
Top