What are the main approaches for business time series forecasting? Explain how a linear regression model can be set up to capture a time series with a trend and/or seasonality?

Assignment Question

Discussion: What are the main approaches for business time series forecasting? Explain how a linear regression model can be set up to capture a time series with a trend and/or seasonality? Provide examples of organizations that can benefit from this? Also reply to two of your peers, Peer 1 (Marie) Here are my thoughts on this week’s discussion.

The main approach for business time series forecasting is the measurement of units of time. This may be measured by payroll/hourly, sales that are daily or monthly. If it is a fixed time this methodology would show how patterns are identified whether past, present, or future. Some of the components may be seasonal, irregular, cyclical, or even today’s current markets and trends. (Shmueli, Bruce, Gedeck, & Patel, 2019). For example, inventory control could be considered as the trend component whereas the patterns are usually analyzed over time by costs either that have increased or decreased. Trend forecasting can be used in supply chain management in many ways. It can entail forecasting shipments or inventory conditions for generating information for ordering or planning production. It should be noted that shipments are a tracked variable while inventory represents a tracking variable. (Saeed, 2008). Overall, the process for inventory modeling for supply chain management is to forecast the costs of the product and delivery methods for future sales. Which for lack of better words is inventory forecasting based on supply and demand. The business would have better revenue when utilizing this data. Since we are in the holiday season right now, Amazon is tracking data at an accelerated rate and the inventory management platform would need a team to operate for possible out of stock items and or back orders. Therefore, there must be accurate inventory forecasting in place to fill the warehouses and historical data from last year would only benefit Amazon to anticipate this seasonal sales trends. Of course, demographics and geographical location will need to be another factor in buying behaviors, but I think using linear regression modeling you can correlate the time quite efficiently.

The textbook gave a notable example for retail sales in a time plot diagram for quarterly sales over a 6-year period in which the output from the regression model would be the same for supply chain management such as Amazon during the holiday season. (Shmueli, Bruce, Gedeck, & Patel, 2019). In conclusion, depending on the data mining analysis the business would need to choose the right forecasting model and technique to have the best accuracy without errors for a better economic prediction. References Saeed, K. (2008). Trend Forecasting for stability in supply chains. Journal of Business Research. Shmueli, G., Bruce, P. C., Gedeck, P., & Patel, N. R. (2019). Data Mining for Business Analytics. Wiley. Peer 2 (vickie) Time series forecasting is a method of predicting future outcomes by analyzing historical data that gives businesses insight into trends hidden in the data. To use time series forecasting effectively, the historical data to be input must be accurate as well a capture a sufficient period of time. Even when these conditions are met, the output would suffer accuracy should the forecast be projected too far into the future. Historical data used in this method is typically data captured in a timestamp format and contains the year, month, day, hour, minute, the event that occurred, much like a video of security feed. The analyst must choose a starting point with which to create a unit of measure and the most accurate prediction made will be the next unit in the future. Depending on the data and the specificity needed, units can be measured in days, hours, weeks, etc. Businesses can predict sales a month out, meteorologists can predict the weather down to hours or minutes, and farmers can make seasonal predictions. A time series regression model analyzes a time series by associating a linear relationship between the output variables and the input variables.

To use a time series regression model, historical data should be analyzed to discover any trends, once established, a line chart is created from the analyzed time series. The line chart will display any changes of an independent variable during the captured period. Prices for agricultural commodities fluctuate based on supply and demand but there are other things that can influence prices. Weather, fuel costs, and economic demand. Time series forecasting can point out trends in these areas and allow farmers and agricultural officials to predict how much produce needs to be grown, what price allows farmers to recoup operating costs, what customers are willing to spend and even when things should be planted. Burba, D. (2023, July 31). An overview of time series forecasting models.

Medium. https://towardsdatascience.com/an-overview-of-time-series-forecasting-models-a2fa7a358fcb Yoo, T.-W., & Oh, I.-S. (2020, November 18). Time series forecasting of Agricultural Products’ sales volumes based on seasonal long short-term memory. MDPI. https://www.mdpi.com/2076-3417/10/22/8169