In: Statistics and Probability
A financial planner tracks the number of new customers added each quarter for a 6 year period. The data is presented below:
Year Quarter New Year Quarter New
2014 I 31 2017 I 69
II 24 II 54
III 23 III 46
IV 16 IV 32
2015 I 42 2018 I 82
II 35 II 66
III 30 III 51
IV 23 IV 38
2016 I 53 2019 I 91
II 45 II 72
III 39 III 59
IV 27 IV 41
Create a simple linear trend regression model. Let t=0 in 2013: IV. This is a computer deliverable.
(a) Interpret the slope coefficient.
(b) Test to see if the number of new customers is increasing over time. Use alpha = 0.01.
(c) Test to see if the model has explanatory power. Use alpha = 0.05.
(d) Forecast the number of new customers in the first and second quarters of 2020.
Create a multiple regression equation incorporating both a trend (t=0 in 2013: IV) and dummy variables for the quarters. Let the first quarter represent the reference (or base) group. Complete (e) thru (h) using your results. This is a computer deliverable.
(e) Test to see if there is an upward trend in new customers. Use alpha = 0.01.
(f) Test to see if the model has explanatory power. Use alpha = 0.05.
(g) Forecast the number of new customers in the first and second quarters of 2020.
(h) Test for the existence of first order autocorrelation, use alpha = 0.05. The calculated dw = 1.19.