trend removal

Page numbers are for draft copy at present; they will be replaced with correct numbers when final book is formatted. Chapter numbers are correct and will not change now.

Trend removal (or trend adjustments) try to separate long terms trends from shorter term fluctuations in data. Typically the long-term trend is estimated, often as a linear growth or decay and then subtracted from the data values. The term is usually applied to time series data, but can be appled equally to other forms of sequential data. Some transformations implicitly remove trend. For example taking successive differences of a time series (t_i+1−t_i) effectively removes any linear trend. Simlarly, taking successive ratios (t_i+1/t_i) removes exponential trend.

Defined on page 318

Used on Chap. 14: pages 318, 333, 334, 336; Chap. 23: page 571

Also known as de-trending, trends, trend adjustments