Kokonaisvirheen

Kokonaisvirheen is a Finnish term used in measurement theory and statistics to describe the total deviation of a reported value from the true value. In practical use, it represents the overall error present in a measurement or estimate.

A common way to understand kokonaisvirhe is through its decomposition into two main components: systematic error (bias) and random error (statistical fluctuations). The total error can be thought of as the sum of these sources: e_total = e_systematic + e_random. Systematic error is a reproducible bias that shifts all measurements in a particular direction, while random error varies from measurement to measurement due to unpredictable factors. Parsing kokonaisvirhe into these parts helps in identifying how to improve accuracy: calibration and method improvements reduce systematic error, while increasing measurement repetitions reduces the impact of random error.

In formal terms, the observed value is often modeled as X_hat = X + e_total, where X is the true value and e_total encompasses both bias and variability. For many analyses, the expected squared total error (the mean squared error) can be expressed as MSE = Bias^2 + Variance, highlighting the contributions of systematic and random components.

The term is used in Finnish scientific and technical literature, including metrology, experimental physics, engineering, and econometrics. Its genitive form, kokonaisvirheen, appears in phrases such as kokonaisvirheen suuruus (magnitude of total error) and kokonaisvirheen arviointi (assessment of total error).