Question

Part B Time-series analysis and forecasting [4+8+6+6+22+9= 55 Marks]

This part you will be doing time-series analysis and forecast a time-series variable. Gather data for any country for any ONE variable with at least 20 periods, the data to be quarterly. This part to be done in Excel file only. Provide below information:

Country

Variable Variables

measurement units

Variable simplified measurement units

# of periods ( n ? 20)

Note: Variable (simplified measurement units) will be used in the analysis.

1. Smoothing Technique: To produce a better forecast, we need to determine which components are present in a time series. To identify the components present in the time series, we need first to remove the random variation. We can reduce random variation by smoothing the time series. EC203 Assignment Semester 1 2018 Page 3 of 4

i. Using exponential smoothing technique, construct new series of your selected variable. [Select appropriate value of ? = Smoothing constant where 0 ? ? ? 1, and justify your choice.]

ii. Plot a time-series chart of your selected variable with original values and values found after exponential smoothing on a same graph and interpret.

Answer 1

Country: India

Variable: Inflation WPI (inflation wholesale price index)

Variable Unit: Indian currency Rs. (INR)

The data is available from 1996Q1. It is quarterly frequency data.

Below is presented a screenshot of smoothing technique using the exponential method. The value of alpha lies between 0.1 to 0.9. The appropriate value of alpha is chosen by lowest mean square error. Each work is shown in the excel file attached. The appropriate alpha here is 0.9.

1. 

ii.