Need help now? Support now: crisis lines by country.
Pain words · Mathematical Institute for Research of Pain
Noise
Also called random error
In statistics, noise is the part of the variation in measurements that a chosen model leaves unexplained; it says nothing about whether that variation matters.
In plain words
Rate pain at the same time each day and the numbers move even on days when nothing seems to have changed. A statistical model splits each rating into a part it explains, such as a slow trend or a weekly pattern, and a leftover part, which it calls noise. Calling the leftover “noise” is a statement about the model. A different model, one that includes sleep or weather, might explain some of the same movement.
What researchers mean
In the simplest additive model an observation is written as a signal plus an error with mean zero. White noise is error that is uncorrelated between any two different times and has constant variance; other kinds of noise carry memory from one time to the next. In pain research the day-to-day movement of ratings is also studied as a quantity in its own right: in two diary studies, Schneider and colleagues (2012) found substantial day-to-day variability, and people differed considerably, and somewhat reliably, in how much their ratings moved. Diary research also separates differences between people from changes within one person over time; Stone and colleagues (2004) found that weekly recall and averaged momentary ratings agreed moderately to well between people but poorly within a person.
Models often assume that the noise on one day is independent of the noise on the next. When bad days cluster together, that assumption makes the uncertainty of the estimates look smaller than it is.
Noise in this sense does not mean that the changes are unreal or unimportant, or that anyone is reporting them wrongly.
The maths
An additive model splits each observation into two parts:
- rating on day t
- the part the model explains
- the rest: mean zero
Which part counts as noise depends on the model: adding a variable such as sleep to moves some variation out of .
These pages explain words, reasoning and evidence. Nothing here is medical advice or a forecast about any person.
Sources
- Schneider S, Junghaenel DU, Keefe FJ, Schwartz JE, Stone AA, Broderick JE. Individual differences in the day-to-day variability of pain, fatigue, and well-being in patients with rheumatic disease: associations with psychological variables. Pain. 2012;153(4):813–822. doi:10.1016/j.pain.2012.01.001
- Stone AA, Broderick JE, Shiffman SS, Schwartz JE. Understanding recall of weekly pain from a momentary assessment perspective: absolute agreement, between- and within-person consistency, and judged change in weekly pain. Pain. 2004;107(1–2):61–69. doi:10.1016/j.pain.2003.09.020
Related words
Questions about this
Inside the Institute
Pain math explains what the Institute does, and Questions about pain answers some questions people in pain ask, in the same layers. The rooms, the Lab and the full Pain Bias Codex open with an invite code.
No code yet? Ask for an invite. A person reads every request and writes back by hand with an invite code, so it may take a few days.
Need help now? Support now: crisis lines by country.
Except where a page says otherwise, the Institute’s own text and figures are shared under CC BY-NC 4.0. Pictures and quotations from others keep their own credits and licences.