**Understanding Moving Average and the Formula Of Moving Average** – Moving Average is one of the simple business forecasting methods and is often used to estimate conditions in the future by using data sets- past data (historical data).

In Operations and Production Management, the data set here can be in the form of sales volumes from the company’s history.

The time period of the data set can be in the form of Annual, Monthly, Weekly or even Daily. This *Moving Average* Forecasting Method is often used in business forecasting such as forecasting market *demand* ( *demand forecasting)*), technical analysis of stock and forex movements as well as estimating business trends in the future.

## Definition of *Moving Average* and How to Calculate it

Basically, the Definition of *Moving Average* is a forecasting method that calculates the average value of a coherent time and then is used to estimate the value in the next period.

*Moving Average** is* obtained through the addition and search for the average value of a certain number of periods, then eliminate the longest value and add new values.

This *Moving Average* method is better used to calculate data that is stable or data that does not fluctuate sharply (data that changes up and down is very drastic). This is because the data in each period are given the same weight so that they cannot represent certain periods of a specific nature or the data of the last period which is usually rated as the best data in describing the current conditions.

Therefore, other *Moving Average* Methods emerge to try to overcome them, the other *moving average* methods include *the Weighted Moving Average* or abbreviated as WMA and the *Exponential Smoothing* Method.(Leveled Smoothing Method). While this simple *Moving Average* method is often referred to as *Simple Moving Average* or abbreviated as SMA.

### Formula *Moving Average* (Moving Average formula)

The *Moving Average* formula is as follows:

**MA = ΣX / Number of Periods**

Note:

**MA** = *Moving Average*

**ΣX** = Overall Sum of all time period data taken into account

**Number of Periods** = Number of moving average periods

or can be written with:

**MA = (n1 + n2 + n3 + …) / n**

Note:

**MA** = *Moving Average*

**n1** = first period data

**n2** = second period data

**n3** = third-period data and so on

**n** = Number of moving average periods

### Case and How to Calculate *Moving Average* (Moving Average)

ABC, which is engaged in the manufacturing of cellphones, would like to forecast sales of cellphones for April and May using its monthly data starting from January. The moving period is 3 months. The following are the methods and results of the calculations.

Month | Sales (unit) | Estimated (unit) |

January | 22,500 | – |

February | 37,500 | – |

March | 30,000 | – |

April | ? | |

May | ? |

**The solution:**

Estimated Sales for April are:

_{April} MA = (22,500 + 37,750 + 30,000) / 3 _{April}

MA = 90,000 / 3 _{April} MA = **30,000**

So the estimated cellphone sales in April are around **30,000 units.**

We can continue for May by using the estimated data or by waiting for the actual results for the month. For example, the actual data obtained in April was 35,000 units, then the calculation is as follows:

MA _{May} = (37,500 + 30,000 + 35,000) / 3

MA _{Mel} = 102 500/3

MA _{May} = **34 167**

With these calculations, it was found that the estimated cellphone sales for May were around **34,167 units** .

Note: For the May calculation, January sales are eliminated and replaced with April sales results. This is because the calculation of our Moving Average is 3 months.

We can create a sales forecast table with a table like the following:

Month | Sales (unit) | Estimated (unit) |

January | 22,500 | – |

February | 37,500 | – |

March | 30,000 | – |

April | 35,000 | 30,000 |

May | ? | 34,167 |

We can continue this table after getting the actual sales data. The following are examples of tables and graphs for forecasting or forecasting sales and actual sales data.