Real-world examples of moving averages in time series analysis

Starting with real examples of moving averages in time series analysis

Before definitions and formulas, it’s more helpful to see how moving averages show up in the wild. Some of the best examples of moving averages in time series analysis come from places you probably check every day:

• A 7‑day rolling average of new COVID‑19 cases on a public health dashboard.
• A 50‑day and 200‑day moving average on a stock chart.
• A 12‑month moving average of inflation or unemployment.
• A 30‑day moving average of website visits in Google Analytics.
• A 3‑month moving average of retail sales to smooth seasonal spikes.
• An exponential moving average of heart rate data from a fitness tracker.

Each one is a concrete example of taking a jagged, noisy time series and replacing each point with an average of nearby points. That simple trick makes it much easier to see patterns, turning chaos into something you can actually interpret.

Finance: classic examples of moving averages in time series analysis

Financial markets are probably the most famous examples of examples of moving averages in time series analysis. Traders don’t just like moving averages; they build entire strategies around them.

Stock prices and trend-following

Take a daily closing price series for a stock like Apple or an index like the S&P 500. The line jumps up and down every day. To see the underlying trend, analysts often overlay:

• A 20‑day simple moving average (short-term trend)
• A 50‑day simple moving average (medium-term trend)
• A 200‑day simple moving average (long-term trend)

A simple moving average (SMA) takes the most recent k days, adds the prices, and divides by k. For a 20‑day SMA, today’s value is the average of the last 20 closing prices.

Two of the best examples in practice:

• Support and resistance: When the price falls toward the 200‑day SMA and bounces, traders say the moving average is acting as support. When it rises to the 200‑day SMA and stalls, it’s seen as resistance.
• Golden cross / death cross: When the 50‑day SMA crosses above the 200‑day SMA, many traders interpret that as a bullish shift (golden cross). The reverse is the bearish death cross.

These are textbook examples of moving averages in time series analysis being used not just to summarize data but to trigger decisions: buy, sell, or stay out.

Exponential moving averages in algorithmic trading

Algorithmic traders often prefer the exponential moving average (EMA), which gives more weight to recent prices. The EMA reacts faster to new information, which matters when you’re trading intraday.

A common configuration:

• 12‑period EMA and 26‑period EMA for short- and medium-term trends
• 9‑period EMA applied to the difference between them (the MACD signal line)

That combination underpins the MACD indicator, one of the best-known real examples of moving averages in time series analysis in technical analysis. It’s a reminder that once you understand how to compute a moving average, you can stack, subtract, and transform them into more advanced tools.

For a deeper dive into financial time series concepts, the Federal Reserve’s education resources on data and indicators are a good starting point: https://www.federalreserve.gov/education.htm

Public health: 7‑day rolling averages for noisy case counts

If you watched COVID‑19 dashboards during the pandemic, you’ve already seen real examples of moving averages in time series analysis used at national scale.

Smoothing daily COVID‑19 cases and deaths

Daily reported cases and deaths are extremely noisy:

• Fewer tests and reports on weekends
• Reporting backlogs after holidays
• Occasional data dumps from labs or states

To handle this, organizations like the Centers for Disease Control and Prevention (CDC) publish 7‑day moving averages of new cases, hospitalizations, and deaths. Instead of staring at a roller coaster of daily values, you get a smoother line that reflects the trend over the past week.

You can see this approach in action on CDC COVID data pages: https://covid.cdc.gov/covid-data-tracker/

This is one of the clearest examples of examples of moving averages in time series analysis where the goal is communication. The raw data is still there, but the moving average makes the message understandable for policymakers and the public.

Flu and respiratory illness surveillance

The same logic applies to influenza-like illness (ILI) and other respiratory viruses. Weekly data can still be spiky, especially at the start and end of flu season. Public health analysts often compute 3‑week or 5‑week moving averages to:

• Compare current flu activity to previous seasons
• Detect the onset of a new wave
• Monitor the impact of vaccination campaigns

In these contexts, moving averages are not fancy forecasting tools; they’re practical filters that separate signal from noise.

For more on how health surveillance data is handled, see CDC’s FluView surveillance information: https://www.cdc.gov/flu/weekly/overview.htm

Climate and environment: long-term rolling averages

Climate data is another field packed with real examples of moving averages in time series analysis, especially when the goal is to distinguish long-term trends from seasonal patterns and short-term variability.

Temperature anomalies and climate trends

Daily temperatures bounce around a lot. Even annual averages can be influenced by unusual events like major volcanic eruptions or strong El Niño years. Climate scientists therefore use multi-year moving averages.

Common examples include:

• 5‑year moving average of global surface temperature anomalies
• 10‑year or 30‑year moving average to highlight long-term warming trends

These rolling averages smooth out year-to-year noise and make it easier to see the steady upward trend associated with climate change.

Agencies like NASA and NOAA publish graphs that use these kinds of moving averages. For example, NOAA’s climate data portal discusses running means and long-term averages for temperature and precipitation series.

Air quality and pollution monitoring

Air quality indices (AQI) are often reported daily, but regulators may look at 30‑day or 90‑day moving averages of particulate matter (PM2.5) or ozone levels to assess chronic exposure. A single bad day is worrying, but persistent elevation over a 90‑day moving average is a different level of concern.

Here, moving averages support regulatory decisions and health risk communication, not just visualization.

Business, sales, and operations: practical examples include forecasts and staffing

In business analytics, some of the best examples of moving averages in time series analysis are refreshingly down-to-earth. Think:

• Monthly revenue
• Daily website traffic
• Call center volume
• Manufacturing output

Sales and demand forecasting

Retailers often track weekly or monthly sales for thousands of products. The data is messy:

• Promotions spike sales for a week.
• Holidays create predictable peaks.
• Supply chain hiccups create dips.

Analysts might use a 3‑month moving average as a baseline forecast for the next month, especially when they need something simple and transparent. That 3‑month window balances responsiveness (not too stale) with stability (not whipsawed by one weird month).

A common example of moving averages in time series analysis here:

• A grocery chain uses a 4‑week moving average of past sales to set automatic reorder quantities.
• A subscription service uses a 7‑day moving average of new signups to detect marketing campaign impact.

These are not theoretical. They are exactly the kinds of rules you’ll find embedded in spreadsheets and basic inventory systems.

Call centers and staffing

Call centers live and die on forecasting call volume. A typical workflow:

• Compute a 7‑day moving average of calls per hour to get a short-term baseline.
• Overlay a 28‑day moving average to capture broader demand shifts.

If today’s 7‑day average is consistently above the 28‑day average, managers may add staff or extend hours. This is another real example of moving averages in time series analysis driving operational decisions.

Web analytics and digital products: rolling averages for engagement

Digital products generate time series by the second: page views, active users, click-through rates, in-app events. Raw series at that resolution are useless without smoothing.

Website traffic and marketing performance

A marketing team might track:

• Daily sessions and users
• Conversion rate (purchases / sessions)
• Email open and click rates

Instead of reacting to every dip or spike, they compute:

• A 7‑day moving average of sessions to smooth weekday/weekend patterns.
• A 30‑day moving average of conversion rate to judge whether a site redesign actually improved performance.

These are straightforward examples of examples of moving averages in time series analysis where the main job is to reduce overreaction. Without a moving average, it’s too easy to panic over a bad Tuesday and celebrate a random good Friday.

Product metrics and anomaly detection

Product teams monitoring metrics like daily active users (DAU) or retention often rely on exponential moving averages because they want recent changes to matter more.

For instance:

• A product manager tracks a 14‑day EMA of DAU.
• When the current day’s DAU falls more than, say, 3 standard deviations below the EMA, an alert fires.

This is a concrete example of moving averages in time series analysis combined with statistical thresholds to detect anomalies in real time.

Comparing types: simple, weighted, and exponential moving averages

So far, we’ve been mixing different flavors of moving averages. It’s worth pausing to sort them out, using the real examples above as anchors.

Simple moving average (SMA)

The SMA treats each observation in the window equally. It’s easy to compute and explain, which is why it shows up in:

• 7‑day averages of COVID cases
• 50‑day and 200‑day stock price averages
• 3‑month averages of sales

An SMA is a good first choice when:

• You care about a clear, stable trend line.
• You need something stakeholders can understand in one sentence.

Weighted moving average (WMA)

A weighted moving average assigns different weights to each point in the window. Recent values usually get higher weights. For example:

• A demand planner might use weights of 0.6 (last month), 0.3 (two months ago), 0.1 (three months ago).

This is an example of moving averages in time series analysis where you know that more recent data is more informative, but you still want a finite window.

Exponential moving average (EMA)

The EMA is a special weighted average that decays exponentially back in time. It has two big advantages:

• It reacts faster to changes than an SMA with the same nominal window.
• It has a nice recursive formula: today’s EMA is a blend of yesterday’s EMA and today’s value.

Real examples where EMAs dominate:

• Intraday trading signals
• Real-time monitoring (system metrics, app performance, heart rate)

When you see the phrase “smoothed metric” in a dashboard that updates every few seconds, there’s a good chance an EMA is doing the smoothing.

Pitfalls: when moving averages can mislead

It’s easy to treat moving averages as harmless smoothing, but they can distort your view if you’re not careful.

Some examples include:

• Lag: A 30‑day moving average of stock prices will always trail behind a sharp reversal. If you use it to trigger trades, you’ll be late.
• Edge effects: At the beginning of a time series, you don’t have enough history for a full window. Some tools drop those early points; others use partial windows. That can change the story your chart tells.
• Masking change: A 12‑month moving average of unemployment can hide a sudden spike, which is a problem if you’re trying to detect recessions early.

These are still examples of moving averages in time series analysis, but they highlight a key point: smoothing is a trade-off. You gain clarity at the cost of responsiveness.

FAQ: short answers about examples of moving averages

Q1. Can you give a simple example of a moving average in everyday life?
Yes. A 7‑day average of your daily step count from a fitness tracker is a classic example of a moving average. It smooths good and bad days into a more stable trend.

Q2. What are common examples of moving averages in economics?
Economists often use 3‑month or 12‑month moving averages of inflation, unemployment, or GDP growth. These examples include rolling averages that make it easier to see business cycle patterns without being distracted by one-off shocks.

Q3. Which example of a moving average is best for forecasting sales?
There isn’t a single best example, but in practice many teams start with a 3‑month simple moving average for short-term forecasts, then experiment with weighted or exponential moving averages when they need faster reaction to shifts in demand.

Q4. Are moving averages only for visualization, or can they be used for decisions?
They’re used for both. The golden cross in stock trading, 7‑day averages for public health thresholds, and 7‑day vs. 28‑day averages for staffing decisions are all real examples where moving averages trigger specific actions.

Q5. Where can I learn more about statistical methods for time series?
University statistics and data science departments often publish open course materials. For example, many .edu sites host lecture notes on time series analysis that cover moving averages, autoregressive models, and forecasting in more mathematical detail.