Simple vs Weighted moving average

Building on your earlier reading about forecasting, this page introduces a specific tool called the Weighted Moving Average (WMA).

In simple terms, a WMA is a way to predict the future by looking at the recent past, but giving "extra importance" to what happened most recently.

Breaking Down the Concept

* Simple vs. Weighted: In a simple moving average, every day in your data is treated as equally important. In a weighted moving average, you decide which days matter more.

* The "Recency" Rule: Usually, the most recent data point is assigned the highest weight because it is the most relevant to what might happen tomorrow.

* The Math Rule: To keep the forecast accurate, the sum of all the weights you choose must always equal 1 (or 100%).

Why use a Weighted Moving Average?

Imagine you are forecasting sales for a coffee shop:

* Simple Average: You look at sales from Monday, Tuesday, and Wednesday and just average them.

* Weighted Average: You realize that Wednesday’s sales are a better hint of Thursday’s sales than Monday's were. So, you give Wednesday a weight of 0.5, Tuesday 0.3, and Monday 0.2.

By doing this, your forecast "reacts" more quickly to new trends or changes in the market compared to a simple average.

Comparison at a Glance

| Feature | Simple Moving Average | Weighted Moving Average |

|---|---|---|

| Weighting | Every data point is equal. | You choose the importance of each point. |

| Focus | Better for very stable, unchanging data. | Better for data where the latest info is most vital. |

| Calculation | Sum of items / Number of items. | (Value 1 \times Weight 1) + (Value 2 \times Weight 2)... |

Would you like me to walk through a practice math problem using this method, or would you like to see how this compares to "Exponential Smoothing"?