Question

a) Discuss what the time series decomposition tells you about your data series. Include discussion of the seasonal, cyclical, and trend components.

b) Compare the time series decomposition forecasts with Holt Winters. Within the sample, is the times series decomposition or Holt Winters more accurate? Try to explain why. (see below for data)

Audit Trail- Statistics Accuracy Measures MAPE R-Square Value 1.65% 98.99% Forecast Statistics Mean Standard Deviation Value 5.06 1.04 Method Statistics Method Selected Basic Method Decomposition type Value Trend (Linear) Regression Multiplicative Components of Decomposition Forecasted Data Centered Moving Average Seasonal Indices Date Original Data CMA Trend Cycle Factors 000-0 0 0 0 0 0 0 0 0 001100 308529 221110 07 539-5 4 1397407 333-3 3 3 3 3 3 3 3 2 9999900 01 20 2 2-2 9 01 20 2 MAM Ju Ju A SON De Ja Fe

Answer 1

Two ways to increase the statistical power of an experiment overlap in the two sampling distributions for null hypothesis and alternative hypothesis Larger the sample size, the higher the power 2. Decrease random error - Reduce the variance: smaller variance yields higher power. Power is higher when the standard deviation is smaller than when it applied consistently.