Moving Average Calculator | SMA & EMA
Calculate simple moving average (SMA) and exponential moving average (EMA) for time series data.
Paste or type numbers separated by commas
Equivalent period ≈ 5.7
Show
What Is the Moving Average Calculator | SMA & EMA?
This Moving Average Calculator computes SMA, EMA, and WMA simultaneously for your time series data, with crossover detection (golden cross / death cross) between the SMA and EMA. Paste your data as comma-separated values and instantly see smoothed trends alongside the raw series.
- ›Three MA types simultaneously: toggle SMA, EMA, and WMA columns on/off to compare how each smooths your data differently.
- ›Golden / death cross detection: the calculator scans each data point and flags where the SMA crosses above (golden) or below (death) the EMA.
- ›EMA equivalent period: each α value is converted back to an equivalent SMA period (2/α − 1) for easy comparison.
- ›Trend detection: compares the last SMA value against the average of the last few SMA values to classify trend as rising, falling, or flat.
- ›Weighted Moving Average: WMA assigns linearly increasing weights, period n gets weight n, period n-1 gets weight n-1, etc., so recent data matters more than SMA but with linear rather than exponential decay.
Formula
| Type | Formula | Key Property |
|---|---|---|
| SMA(n) | Sum of last n values ÷ n | Equal weight to each period; lags the most |
| EMA(t) | α × x(t) + (1−α) × EMA(t−1) | Exponential decay; recent values weighted more |
| α (smoothing) | 2 / (n + 1) | EMA with n periods uses α = 2/(n+1) |
| Equiv. period | 2/α − 1 | Convert α to "equivalent SMA period" |
| WMA(n) | Σ wᵢxᵢ / Σwᵢ, wᵢ = i | Linear weights; newest period has weight n |
| Golden cross | Short-term MA crosses above long-term MA | Bullish signal in technical analysis |
| Death cross | Short-term MA crosses below long-term MA | Bearish signal in technical analysis |
How to Use
- 1Enter your data as comma-separated numbers (e.g. 10, 12, 11, 14, 16, 15, 18).
- 2Set the period, this defines how many past values each MA point uses.
- 3For EMA, set the smoothing factor α (default 2/(n+1), or enter a custom value between 0 and 1).
- 4Toggle the SMA, EMA, and WMA checkboxes to show or hide each column.
- 5Click Calculate (or press Enter) to generate the full moving-average table.
- 6Review the crossover alerts below the table for golden/death cross signals.
- 7Click Clear to reset all inputs and start fresh.
Example Calculation
Example, 5-period SMA, EMA (α=0.333), WMA
Golden/Death Cross Example
Choosing between SMA, EMA, and WMA
- ›SMA: best for smoothing noisy data where all periods are equally important (e.g. inventory levels)
- ›EMA: best when recent data is more predictive than old data (e.g. stock prices, live metrics)
- ›WMA: intermediate, more responsive than SMA but less jumpy than EMA; good for short-term forecasting
Understanding Moving Average | SMA & EMA
Simple Moving Average (SMA) Explained
The Simple Moving Average is the arithmetic mean of the last n data points. SMA(5) at day 10 = (day6 + day7 + day8 + day9 + day10) / 5. As each new data point arrives, the oldest drops off and the newest is added, the "rolling window" moves forward. SMA gives equal weight to every period in the window, making it easy to interpret but slow to react to sudden changes.
SMA is used extensively in technical analysis (50-day and 200-day SMAs are benchmarks for long-term trend), quality control (X-bar control charts), economics (rolling unemployment rates), and time-series forecasting as a baseline model. A longer period produces smoother output with more lag; a shorter period tracks the data more closely but retains more noise.
- ›SMA(n) needs n data points before the first value can be computed.
- ›All n data points have equal influence, a spike from 5 periods ago affects today's SMA.
- ›SMA "forgets" old data abruptly when a value exits the window.
- ›Cross of a short-period SMA above a long-period SMA is a classic bullish technical signal.
Exponential Moving Average (EMA) and the Smoothing Factor α
The EMA is a recursive weighted average where the weight of each past observation decreases exponentially. The smoothing factor α ∈ (0,1) controls how quickly old values lose influence. EMA(t) = α × x(t) + (1−α) × EMA(t−1). A high α (e.g. 0.8) tracks recent changes quickly but retains noise. A low α (e.g. 0.1) produces a very smooth curve that reacts slowly.
The conventional choice is α = 2 / (n+1), which makes the EMA approximately equivalent to an n-period SMA in terms of how far back its "memory" extends. For example, α = 2/(10+1) ≈ 0.182 gives a 10-period equivalent EMA. The reverse formula 2/α − 1 converts any α back to an equivalent period. EMA is preferred over SMA in most financial applications because it weights recent prices more heavily, making it more responsive to new information.
- ›EMA can be computed with a single pass through the data, no need to store the full window.
- ›EMA never fully "forgets" old data, weights decay but never reach zero.
- ›EMA with small α: long memory, very smooth, lags significantly.
- ›EMA with large α (e.g. 0.5): nearly as responsive as raw data.
Weighted Moving Average (WMA), Linear Decay
The Weighted Moving Average assigns linearly increasing weights to observations in the window: the oldest observation gets weight 1, the second oldest gets weight 2, …, and the most recent gets weight n. WMA(5) = (1×x₋₄ + 2×x₋₃ + 3×x₋₂ + 4×x₋₁ + 5×x₀) / (1+2+3+4+5). WMA is more responsive than SMA to recent changes (because recent data is heavier) but less aggressive than EMA, it is a useful middle ground for short-term trend following.
- ›Like SMA, WMA requires n data points and forgets old data sharply when it exits the window.
- ›Unlike EMA, WMA has no smoothing parameter to tune, only the period n.
- ›WMA is often used in Hull Moving Average (HMA), which combines WMA at different periods to reduce lag.
- ›WMA is preferred in some algorithmic trading strategies for its balance of smoothness and responsiveness.
Golden Cross and Death Cross Signals
A golden cross occurs when a shorter-period moving average crosses above a longer-period moving average, historically a bullish long-term signal. The most watched instance in equities is the 50-day SMA crossing above the 200-day SMA. The opposite, the death cross (50-day SMA crossing below 200-day SMA), is considered bearish. These crossovers are lagging indicators, they confirm a trend has already begun rather than predicting it.
This calculator detects crossovers between the SMA and EMA computed at the same period, useful for shorter-term data series. Each point in the table where the SMA transitions from below the EMA to above it (or vice versa) is flagged as a golden or death cross respectively. For multi-period crossover analysis (e.g. 50-day vs 200-day), compute two separate table instances with different periods and compare the final values manually.
Frequently Asked Questions
What is the difference between SMA and EMA?
Both are moving averages that smooth time-series data, but they weight historical observations differently:
- ›SMA: equal weight to all n observations in the window. Simple, transparent, more lag.
- ›EMA: exponentially decreasing weights, most recent observation has the largest weight, then decays by factor (1−α) each period. More responsive, less lag.
- ›For the same period, EMA reacts faster to new data than SMA.
- ›EMA is generally preferred for financial price data; SMA is common in economic statistics and quality control.
Choose SMA when you want a straightforward average with equal historical weight. Choose EMA when recent observations should matter more than older ones.
How do I choose the right period (n) for a moving average?
Period selection involves a bias-variance tradeoff:
- ›Short period (e.g. n=3–5): tracks the data closely, low lag, but retains more noise.
- ›Long period (e.g. n=20–200): very smooth, but reacts slowly to real trend changes.
- ›For financial charts: 9, 12, 20, 50, 100, 200 are standard periods.
- ›For short-term forecasting: try 3–7; compare RMSE of different periods on historical data.
A practical rule: plot both a short-period and long-period MA together. Their relative positions and crossovers often give more information than either alone.
What is a golden cross and does it reliably predict price increases?
A golden cross occurs when a short-period moving average crosses above a long-period moving average. The most famous instance is the 50-day SMA crossing above the 200-day SMA on a stock or index chart.
- ›Golden cross: short MA crosses above long MA → historically bullish signal.
- ›Death cross: short MA crosses below long MA → historically bearish signal.
- ›As a lagging indicator, it confirms a trend that has already started, not a leading predictor.
- ›Academic research finds mixed evidence for reliable excess returns after golden crosses.
Golden/death crosses are most useful as confirmation signals alongside other analysis, not as standalone trading rules. They work better in trending markets and generate more false signals during sideways (ranging) conditions.
What does the EMA smoothing factor α mean and how do I choose it?
The smoothing factor α ∈ (0,1) controls how quickly the EMA responds to new data vs retaining memory of past values:
- ›α = 0.5: half the weight is on the current value, half on the past EMA. Very responsive.
- ›α = 0.1: 10% on current, 90% on past, very smooth, slow to react.
- ›Default convention: α = 2/(n+1) makes EMA "equivalent" to an n-period SMA.
- ›Higher α → lower lag, more noise. Lower α → more smoothing, more lag.
How is Weighted Moving Average (WMA) different from EMA?
Both WMA and EMA give more weight to recent observations, but the decay pattern differs:
- ›WMA: linear weights (1, 2, 3, …, n). Sharp cutoff, data outside the window has zero weight.
- ›EMA: exponential weights that never reach zero, old data always has some (tiny) influence.
- ›WMA responds more uniformly to all data within the window; EMA weights decay faster.
- ›WMA has no free smoothing parameter (only n); EMA has both n and α.
WMA is often used as a building block for more sophisticated indicators like the Hull Moving Average (HMA), which applies WMA to WMA results to further reduce lag while maintaining smoothness.
Can moving averages be used for forecasting future values?
Moving averages are primarily descriptive (smoothing past data) and directional (identifying current trend direction), but they can be used for naive forecasting:
- ›Naive MA forecast: predict next value = current MA value (works well for near-random-walk series).
- ›Double exponential smoothing (Holt): extends EMA to capture trends and make multi-step forecasts.
- ›Triple exponential smoothing (Holt-Winters): adds seasonality, commonly used in demand planning.
- ›Moving averages form the basis of ARIMA models (autoregressive integrated moving average).
For serious forecasting, use MAs as a baseline to beat rather than the final model. More sophisticated methods (ARIMA, Prophet, neural networks) generally outperform simple MA forecasts on structured series.
Why does EMA start with the first data point rather than the first n-period average?
There are two common EMA initialisation methods:
- ›"Seeded" EMA: initialise EMA with the SMA of the first n periods, then apply the recursive formula. This is the most common method in financial charting software.
- ›"Cold-start" EMA: start with the very first data point as EMA₁, then apply the formula immediately. Results differ for early points but converge after several periods.
- ›This calculator uses the cold-start initialisation so that EMA values are available from the first data point.
- ›For long datasets, the choice of initialisation method has negligible effect on recent EMA values.
The cold-start approach is preferred when you have limited data and need an EMA estimate as early as possible. For datasets with 30+ points, the initialisation method rarely matters beyond the first few values.