please see the attached file this needs to submit in excel worksheet. Please read the instruction properly if you have any confusion check the FAQ give in the file.
 I would like you to create a spreadsheet that calculates a weighted average forecast, and charts it as an xy plot, where time is on the x axis. The data is as follows: Actual observations at times T1 through T5 are: You should compute the weighted average forecast for Times T1 through T6. The weighting factors are 0.2,0.3, and 0.5, 0.2 applies to the earliest, 0.3 to the middle, and 0.5 to the latest. Explain/justify how you derived the forecast at T1 and T2. Repeat the exercise, but now change the weighting factors. Explain what you were trying to achieve, how successful you were, and describe any limitations you see with technique. Submit your work  as an Excel spreadsheet. FAQ 1. What is the relationship between the time periods and the observed and forecast values? This is a set of time series data. At time T1, you just have one piece of data, the observed value for T1. So when you forecast for time T2 that is all you have. At time T2 you have the observed values at time T1 and time T2, you use these to calculate the forecast for time T3.  And so on and so forth. . . 2. Do I replace the forecast with observed values? No, you want to keep both sets of data, the gap between forecast and observed is at the root of measures of goodness of the forecast. 3. Can I use some form of regression to calculate early results? I would not do this, this is probably a better technique for y on x plots, this is time series data, and I have not seen that technique used. 4. Can I use just some of the weighting values for early results? Just remember that in a weighted average the weights need to sum to 1, or you will bias the result. 5. How do I calculate forecast values for the early time periods? This is covered in the slides/lecture. Look back if you missed it. 6. How can I tell if my new values for weights are an improvement? Firstly what were you trying to achieve? Did you achieve it? Secondly, if you plot the observed values against the forecast values, what do you see? Thirdly, anything you doing statistics ought to be checked with some measure of goodness, such as a correlation coefficient, a standard deviation, a Chi square test etc. Look back at the lecture/slides for some ideas?