R代写, R编程代写代做 R算法代写

日期：2018-04-17 10:08

Question 1

What is your data about (nomore than 50 words)? Produce appropriate plots in order to become familiarwith your data. Make sure you label your axes and plots appropriately. Commenton these. What do you see? (no more than 50 words per plot

Question 2

Would transforming yourdata be useful (no more than 50 words)? Compare different transformationsgraphically. Choose the best transformation if you think a transformation isneeded. Justify your choice (no more than 50 words).

Question 3

Split your data intotraining and test sets. Leave the last two years’ worth of observations as thetest set. Plot these on the same graph to make sure you have done thisproperly.

Question 4

Apply the two mostappropriate benchmark methods on the training set. Generate forecasts for thetest set and plot them on the same graph. Compare their forecasting performanceon the test set. Which method would you choose as the appropriate benchmark? Justify your answer (no more than 5 lines).(Hint: it will be useful to tabulate your results.)

Question 5

For the best method, do aresidual analysis. Comment on these. What did your forecasting method miss? (nomore than 50 words)

Question 6

Generate point forecasts forthe next 2 years (2016-2017) from the benchmark method you considered best andplot them. Comment on these (no more than 50 words).

Question 7

Can you invent a betterforecasting method for your data? Simply give a brief description of your newmethod - no programming needs to be performed here (no more than 100 words ).This question requires you to think about and only use tools you have acquiredso far in this unit. Only materials from you current toolbox can be used (nodecomposition or exponential smoothing or anything else is allowed). Ifyou think you cannot invent a better method than the four benchmark methods youwill need to justify your answer.