Question

1. CoastCo Insurance is interested in developing a forecast of larceny thefts in the United States.

Year	Larcency Thefts*
1	8,151
2	8,348
3	9,263
4	9,978
5	10,271
6	9,906
7	9,983
8	10,578
9	11,137
10	11,194
11	11,143
12	10,713
13	10,592
14	10,926
15	11,257
16	11,500
17	11,706
18	11,872

1. Plot this series in a time series plot and make a naive forecast for years 2 through 19.
2. Plot actual and forecast values of the series for the years 1 through 19. (You will not have an actual value for year 19 or a forecast value for year 1.)
3. Calculate the RMSE and MAD for years 2 through 18. On the basis of these measures and what you see in the plot, what do you think of your forecast? Explain.

Answer 1

Plot for times series :

Forecast Table :

NOTE: In Naive forecasting, the actual value of previous time period becomes forecasted value for next time period

Observing the forecast graph, we can see that naive forecasting looks effective barring the fact that the predictions are lagged by a time period.

We must observe that we have periods of positive & negative forecast. like periods 2-5 see higher difference in the actual and forecast value.

The forecast is satisfatory but the MAD and RMSE values could be reduced further by applying other forecasting methods like moving averages or advanced techniques like ARIMA.