Can a stock series be daily?

Yes. Frequency does not change the point-in-time nature of each observation.

Can stocks be added across locations?

Yes, when definitions and times match and units do not overlap.

Is stock growth the same as sales?

No. Growth is a change in a level; sales are usually an activity measure over a period.

AI Glossary / Statistics and probability / Beginner

Stock series

Definition

A stock series is a sequence of measurements of a quantity at specified points in time. Examples include inventory on hand at each month-end or the number of active accounts at each reporting date. Each observation describes a level at a moment, rather than activity accumulated during an interval. Distinguishing stock series from flow series is essential when aggregating data, interpreting changes, or building forecasts from operational records.

What it means

A stock can be imagined as a snapshot, while a flow measures what happens over a period. The number of unresolved cases at midnight is a stock; the number of cases opened during the day is a flow. Both can be arranged as time series, but they answer different questions.

The reference point need not always be month-end: it could be the start of a week, a specific time each day, or another defined instant. Consistency matters. A stock series is not necessarily a financial share-price series, although some financial measurements also describe levels at observation times. The statistical meaning here is broader and applies to many business quantities.

How it works

Define the quantity, unit, population, and exact observation time. Apply the same inclusion rules at every point, or document changes that affect comparability. Store the reference date separately from the date on which the report was produced.

Relate changes in stock to relevant inflows, outflows, and adjustments. For a simple count with stable definitions, closing stock equals opening stock plus entries minus exits. Other settings may also involve transfers, reclassifications, corrections, valuation changes, or losses. A stock difference does not necessarily equal one named activity flow.

When converting frequency, choose a summary that fits the question. An end-of-quarter stock may be the last month-end observation, while an average stock requires an appropriate averaging scheme. Summing several snapshots usually counts units repeatedly and does not produce the stock at the end of the longer period.

For forecasting or AI features, respect when each stock observation was actually available. Later revisions and backfilled counts can otherwise introduce information that was unavailable at prediction time. Check that model inputs and targets use compatible reference times and definitions.

Three practical examples

1. A warehouse records units on hand at each daily closing time and separately records receipts and dispatches during each day.

2. A service team monitors unresolved tickets at week-end while tracking newly opened and closed tickets as separate flows.

3. An AI capacity-planning model uses the current number of active jobs together with arrival and completion rates, preserving the distinction between levels and activity.

Think of it this way

Think of water in a tank. The amount in the tank at noon is a stock. Water entering and leaving over the afternoon are flows. Adding the noon and evening water levels does not tell you how much water entered during the day.

Worked example

Consider an original account-count example. A service starts a month with 1,000 active accounts. During the month, 120 accounts become active and 80 leave, with no other changes. Closing stock is 1,000 + 120 − 80 = 1,040 accounts. Net growth is 40 accounts, while gross entries are 120.

Suppose the next two month-end stocks are 1,070 and 1,050. The quarter-end stock is 1,050, not 1,040 + 1,070 + 1,050 = 3,160. The latter sum adds three snapshots and can count the same accounts repeatedly.

The arithmetic average of these three month-end observations is 3,160 / 3, about 1,053.33 accounts. That is a mean of the sampled snapshots, not necessarily the true time-average active-account level during the quarter. Daily changes could make the latter different.

If a later correction removes ten duplicate accounts from the first month-end count, the revised stock becomes 1,030. A reconciliation must identify the correction rather than silently describing all ten as customer exits. The example assumes clear activity definitions; real systems need rules for reactivation, mergers, and other status changes.

Illustrative case study

In a fictional company, Juniper Services adds monthly active-account stocks to produce a quarterly dashboard total. Managers initially read it as the number of customers active at quarter-end. An analyst replaces the total with the final snapshot and adds a clearly labeled average of month-end counts. Entries, exits, and corrections appear in a separate reconciliation. The revised dashboard supports capacity decisions without confusing accumulated activity with a level. It also preserves historical revisions so AI forecasting tests can use the information actually available at each date.

Common misconceptions and mistakes

Summing stocks across time is not generally a meaningful end-period total. However, summing compatible stocks across disjoint units at the same time can be valid, such as inventory across separate warehouses without overlap.

A change in stock is not automatically gross inflow or gross outflow. Different combinations of entries and exits can produce the same net change.

An average of occasional snapshots is not automatically a time-weighted average. Sampling frequency, timing, and within-period variation matter. Changes in scope or valuation can also create movements that are not real additions or removals under the original definition.

Business use

Stock series support staffing, inventory management, backlog monitoring, and capacity planning. They show the state of an operation at comparable moments and help connect activity to resource requirements.

Business reports should label whether a figure is an end-period level, an average level, or a flow over a period. Reconcile major movements and document revisions. When integrating systems, align time zones, cut-off times, and inclusion rules before comparing counts. These details prevent apparently inconsistent dashboards and help AI models learn from measures with stable meanings.

Check your understanding

Answer: A queue has 50 cases at opening, receives 20, and completes 15, with no adjustments.

What is closing stock?

Answer: Fifty-five cases. The 20 arrivals and 15 completions are flows, while 50 and 55 are stocks.

Related terms

Time series
Flow series
Aggregation
reference period
inventory
backlog
net change
data revision

Learn next

Study stock-flow reconciliation and temporal aggregation. Then examine how reporting cut-offs and revisions affect forecasting evaluation.

Aggregation

Sources

Australian Bureau of Statistics, Statistical terms and concepts glossary, Stock series: https://www.abs.gov.au/statistics/understanding-statistics/statistical-terms-and-concepts/statistical-terms-and-concepts-glossary . The primary definition supports the point-in-time interpretation. The account reconciliation, aggregation calculations, and fictional dashboard case are original teaching examples.

Previous: Stock sampling Next: Stoer-Wagner minimum cut