1. Select a dataset with a variable you would like to forecast. You may use a different source other than the ones listed above (be sure to reference the website). 2. State the variable you are forecasting. 3. Select at least eight consecutive data values. 4. Using the Time Series Forecasting Templates, determine the following for the selected variable: ○ moving average, ○ weighted moving average, and ○ exponential smoothing (see the video in the Module 5 LiveBinder for additional information). 5. Copy/paste the results of each method into your post. Be sure to state: ○ the number of periods used in the moving average method. ○ the weights used in the weighted moving average. ○ the value of α used in exponential smoothing. 6. Clearly indicate the “next period” prediction for each method. 7. Choose one of the following: ○ sentence that identifies the prediction. ○ Circle, draw, etc. on the chart to indicate which value is the prediction for the next time period. 8. Suppose that the forecasting results are from three different branches of a company. ○ Based on the MAD (mean absolute deviation) value, how would you prioritize the need to update the forecasting
| Month | Year | Month | Year | |||
| January | 2020 | 50 | January | 2021 | 39 | |
| February | 2020 | 52 | February | 2021 | 50 | |
| March | 2020 | 65 | March | 2021 | 61.5 | |
| April | 2020 | 67.5 | April | 2021 | 71.5 | |
| May | 2020 | 76.5 | May | 2021 | 76 | |
| June | 2020 | 79.5 | June | 2021 | 82.5 | |
| July | 2020 | 83 | July | 2021 | 81 | |
| August | 2020 | 81.5 | August | 2021 | 83.5 |
In this analysis, we will be forecasting monthly sales data for a company over two consecutive years, specifically from January 2020 to August 2020 and January 2021 to August 2021. The goal is to use different time series forecasting techniques to predict sales for the next period, assess their accuracy, and determine the need for updating forecasting methods based on the Mean Absolute Deviation (MAD).
Data Source: The provided data represents monthly sales for the year 2020 and 2021.
Now, let’s proceed with the analysis using three different time series forecasting methods:
Moving Average Method: For this method, we will use a 3-period moving average.
January 2021 Prediction: (39 + 50 + 61.5) / 3 = 50.83 (rounded to 2 decimal places)
Weighted Moving Average Method: We will assign weights of 0.3, 0.4, and 0.3 to the last three months, respectively.
January 2021 Prediction: (39 * 0.3) + (50 * 0.4) + (61.5 * 0.3) = 50.05 (rounded to 2 decimal places)
Exponential Smoothing Method: For exponential smoothing, we will use the alpha (α) value of 0.2.
January 2021 Prediction: (39 * 0.2) + (50 * 0.8) = 47.2 (rounded to 1 decimal place)
Now, let’s prioritize the need to update the forecasting method based on the MAD values. MAD measures the average deviation of the forecasted values from the actual values.
To calculate MAD for each method
MAD for Moving Average = (|39 – 50.83| + |50 – 50.83| + |61.5 – 50.83|) / 3 = 10.83 / 3 ≈ 3.61 (rounded to 2 decimal places)
MAD for Weighted Moving Average = (|39 – 50.05|) = 11.05 (rounded to 2 decimal places)
MAD for Exponential Smoothing = (|39 – 47.2|) = 8.2 (rounded to 1 decimal place)
Based on the MAD values, we can prioritize the need to update the forecasting methods as follows:
Exponential Smoothing (MAD = 8.2): This method has the lowest MAD, indicating the closest fit to the actual data. It should be the preferred forecasting method.
Moving Average (MAD = 3.61): Although it has a higher MAD compared to exponential smoothing, it can still provide reasonably accurate forecasts.
Weighted Moving Average (MAD = 11.05): This method has the highest MAD, suggesting that it may need improvement or replacement with a more accurate method.
In conclusion, when forecasting monthly sales based on the provided data, the exponential smoothing method with an alpha (α) value of 0.2 appears to be the most accurate. However, it’s essential to continually monitor and update forecasting methods to ensure they remain effective in predicting future sales trends.
As a renowned provider of the best writing services, we have selected unique features which we offer to our customers as their guarantees that will make your user experience stress-free.
Unlike other companies, our money-back guarantee ensures the safety of our customers' money. For whatever reason, the customer may request a refund; our support team assesses the ground on which the refund is requested and processes it instantly. However, our customers are lucky as they have the least chances to experience this as we are always prepared to serve you with the best.
Plagiarism is the worst academic offense that is highly punishable by all educational institutions. It's for this reason that Peachy Tutors does not condone any plagiarism. We use advanced plagiarism detection software that ensures there are no chances of similarity on your papers.
Sometimes your professor may be a little bit stubborn and needs some changes made on your paper, or you might need some customization done. All at your service, we will work on your revision till you are satisfied with the quality of work. All for Free!
We take our client's confidentiality as our highest priority; thus, we never share our client's information with third parties. Our company uses the standard encryption technology to store data and only uses trusted payment gateways.
Anytime you order your paper with us, be assured of the paper quality. Our tutors are highly skilled in researching and writing quality content that is relevant to the paper instructions and presented professionally. This makes us the best in the industry as our tutors can handle any type of paper despite its complexity.
Recent Comments