Skip to content

Stockertim Dictionary · Technology

Missing value

Also called missing data, NaN, null value, NA

In short

A blank in a dataset where a figure should be; it means the value is unknown, which is not the same as zero.

  • Intermediate
  • Concept
  • Checked

In detail

Real datasets have holes. A company may not report a particular line item, a share may not trade on a given day, a survey respondent may skip a question, or a file may be corrupted in transfer. Software marks such gaps in different ways: an empty cell in a spreadsheet, NULL in a database, NaN or NA in data libraries, and sometimes a code such as minus 999 in older files. Recognising them correctly is the first step in handling them.

Treating a blank as zero is a common and serious mistake. A missing revenue figure averaged as zero drags the average down, and a missing price read as zero creates a fake 100% loss. The right treatment depends on why the value is missing. If gaps are random, simply leaving them out of averages may be fine. If they are systematic, such as small firms failing to report in bad years, ignoring them biases results, and more careful approaches are needed.

Common options include deleting incomplete rows, forward-filling time series, filling with a group average or a model's estimate, and adding a flag column so later users know which values were filled. Each choice changes results, so analysts state the approach and check that conclusions survive alternatives.

Worked example

A company's revenue is ₦100m in year one, missing in year two and ₦120m in year three.

  1. Average treating the blank as zero: (100 + 0 + 120) / 3 = 73.3
  2. Average of the known years: (100 + 120) / 2 = 110

Counting the gap as zero understates typical revenue by about a third.

In practice

Statistics offices such as the ONS publish methods for handling non-response in surveys, and data teams in banks and fund managers set rules for gaps in prices and accounts before running models.

Watch out

  • A blank is unknown, not zero.
  • Missing data are often not random; dropping rows can bias results.
  • Different tools count blanks differently in averages and totals.

Learn it properly

Official sources

Facts last checked on 4 November 2026.

Education, not advice: the Stockertim Dictionary explains what terms mean. It never recommends buying or selling anything.

English