The data below shows revenue (£000) for a company that Lendright is considering funding over various years and quarters.
(a) Use an appropriate moving average to find the trend component (T).
(b) Plot the original data and your trend values on a scatter diagram.
(c) Find a ‘least squares line’ to best fit your trend values.
(d) Estimate the seasonal variation component (S).
(e) Forecast future revenue, taking both trend and seasonal variation into account in:
(i) quarter 1, year 2014;
(ii) quarter 3, year 2015.
(f) How confident can you be in these forecasts?
Q36:
The spreadsheet with instructions can be found on the instructor’s website, www. pearsoned.co.uk/wall, and provides a template for calculating the components of time series data. Although the layout here is for quarterly data, the concept can be adapted for any time series data. As you will see, you are using the technique of a four quarter centred moving average to estimate the trend of the data (had the data been monthly, you would have used a 12 month centred moving average to estimate the trend). You are also finding the (adjusted) seasonal factors. If you are using your trend values for forecasting you will need to add or subtract the seasonal factors from each quarterly trend estimate to get a more accurate forecast. The quarterly data values shown below are entered on the spreadsheet.
(a) Calculate the trend values.
(b) Calculate the seasonal variation factors.