Also called DFA, a technique to separate out small-scale fluctuations and self-similarity in time-series data, from larger trends that are not stationary. It uses a "modified root mean square analysis" of a random walk.

For example, European currencies may respond to world events such as recession and oil prices in similar ways, compared to dollars or yen, but there may also be smaller linkages or fluctuations between them that the larger trend masks. DFA can detrend the data, i.e. remove the effect of the extrinsic trend.

A mathematical presentation of DFA may be found at

Log in or register to write something here or to contact authors.