Consider the following demand data for a certain product over four quarters in 2019 and four quarter in 2020. (The data can be directly copied into Excel.)

Period	Quarter	Demand
1	2019 Q1	830
2	2019 Q2	1050
3	2019 Q3	930
4	2019 Q4	770
5	2020 Q1	740
6	2020 Q2	930
7	2020 Q3	810
8	2020 Q4	680

We have noticed a seasonal pattern in the data and thus decided to use a cycle decomposition technique with multiplicative seasonality to forecast the trend into 2021, i.e., periods 9 through 12.

a) The seasonal factor for Q3 is [ Select ] ["1.09", "1.03", "0.96", "0.92"] .

b) The deseasonalized demand in Q2 of 2020 (Period 6) is [ Select ] ["791", "974", "1392", "888"] .

c) Based on the linear regression of the deseasonalized data, the deseasonalized forecast for Q4 of 2021 is [ Select ] ["777", "1030", "844", "693"] .

d) The correlation coefficient of [ Select ] ["-0.81", "-0.87", "0.87", "0.81"] for the regression shows a(n) [ Select ] ["decreasing", "increasing"] demand trend and a relatively [ Select ] ["strong", "weak"] fit (predictive power).

e) The forecast including trend and seasonality (FITS) for Q1 of 2021 is [ Select ] ["1016", "1152", "701", "921"] .

f) The four-period moving-average forecast for Q1 of 2021 equals [ Select ] ["945", "790", "995", "886"] and [ Select ] ["overestimates", "underestimates"] the FITS for the same period.