Question 1:[18 Points] The data set souvenirs.csv contains the monthly sales of a shop from January...

Transcribed Image Text:

Question 1:[18 Points] The data set souvenirs.csv contains the monthly sales of a shop from January 1987. (A) Initial examination of the data: Redefine this dataset as a time series. Plot the data along with their ACF and describe the patterns in the graphs. (B) Transformation the data, if necessary: Take logarithms of these data. Does this transformation help the analysis? Why? (C) Initial identification of the dependence orders and degree of differencing: Based on your answer from Part (B), remove any pattern in the time series using regression or differencing. (D) Model suggestion: Fit the following models to the time series resulting from Step (C): 1. AR(1) [order=(1,0,0) in arima' function] 2. MA(1) [order=(0,0,1) in arima' function] 3. ARMA(1,1) [order=(1,0,1) in 'arima" function] Based on the AIC of the model, Which one is a better choice? Use the 'arima' function from 'stats package for this purpose. (E) Do the residual diagnostics for all the models in Part (D). Does your model fit the data well? Use 'sarima' function from 'stats' package for this purpose. Question 1:[18 Points] The data set souvenirs.csv contains the monthly sales of a shop from January 1987. (A) Initial examination of the data: Redefine this dataset as a time series. Plot the data along with their ACF and describe the patterns in the graphs. (B) Transformation the data, if necessary: Take logarithms of these data. Does this transformation help the analysis? Why? (C) Initial identification of the dependence orders and degree of differencing: Based on your answer from Part (B), remove any pattern in the time series using regression or differencing. (D) Model suggestion: Fit the following models to the time series resulting from Step (C): 1. AR(1) [order=(1,0,0) in arima' function] 2. MA(1) [order=(0,0,1) in arima' function] 3. ARMA(1,1) [order=(1,0,1) in 'arima" function] Based on the AIC of the model, Which one is a better choice? Use the 'arima' function from 'stats package for this purpose. (E) Do the residual diagnostics for all the models in Part (D). Does your model fit the data well? Use 'sarima' function from 'stats' package for this purpose.

Answer rating: 100% (QA)

To calculate the internal growth rate IGR we need to determine the retention ratio and return on equ... View the full answer