Answer the first question in 200 words min and reply two the post in a min of 250 words APA format and peer reviewed references for both.
1-In your mind, how does time series differ from regression?
Any thoughts?
2- The major elements in applying the time series regression model to forecasting are the following: seasonal, cyclic and random variation (or noise). When observations of data either increase or decrease over time, then the series is considered to have a trend. There are 3 different types of trend: linear, exponential and s-shape. Linear is when the data increases by the same amount from time period to time period. Exponential trend is when the data increases faster and faster over time. An s-shaped trend is when the data takes a while to increase, but then increases quickly and finally tapers off to a static rate.
Within a time series trend, there can be a seasonal or cyclical component. A seasonal component means that there are several occurrences of a pattern within a year's period. A cyclical component means that there is a cycle to the pattern, but the repeating nature of the pattern is more difficult to predict and can occur over longer periods than a year, meaning they take place over a business cycle rather than a yearly cycle. If there is any irregularity or unpredictable component, than that is considered random variation or noise. This can cause a zig zag appearance in the time series.
Examples of forecasting with time series data could be in forecasting hotel demand during a certain period. If a hotel in Florida is trying to forecast it's occupancy rate for November-February, they may have data that shows higher occupancy rates since it is snow-bird season. Likewise, a NYC hotel may expect higher occupancy during New Year's Eve every year because of it's past trend during that time period. Use of time series analysis allows hotel owners to predict occupancy and then raise or lower their prices accordingly in order to bring in the optimal profit.
Albright, S. C., Winston, W. L., & Albright, S. C. (2015). Business analytics: data analysis and decision making. Stamford, CT, USA: Cengage Learning