Question:

Part I: Time Series Forecasting Models (50%)

You are to build time series forecasting models to predict monthly stock price of your insurance firm (2008, 2009, 2010, 2011, 2012, 2013, 2014, and the first nine months of 2015.

1. Using Minitab 17.2 Time Series: Fitting Process (Single and Double Exponential Smoothing) For each of these two model forms, and for each of the parameter combinations, you need to include a copy of annotated output from Minitab 17.2 (including your data files and drop downs.) 1. Build single exponential smoothing models fitted to the 93 months of stock price of your insurance firm. You need to run a full range of exponential smoothing models with = .1, .2, .3, .4, .5, .6, .7, .8 and .9, for the 9 model runs, and display the measures in a table of fit for each of the models, as well as the optima arima model 10^th model run, and you need to display the following: 1. mean squared error 2. mean absolute deviation 3. mean absolute percent error Then discuss which model gives the best fit over the 93 months period for the stock price of your insurance firm, and the best forecast for months 94, 95, and 96, (October, November, and December 2015), as well as the average of months' forecast (for months 94, 95, and 96). 3. Build double exponential smoothing models fitted to 93 months period for the stock price of your insurance firm. You need to fun a full range of double exponential smoothing models with =.2, .4, .6 and = .2, .4, and .6, as well as the optima arima model 13^th model run. You need to display:1.mean square error2.mean absolute deviation3.mean absolute percent error. Then discuss which model gives the best fit over the 93 month period for the stock price of your insurance firm for months October, November and December 2015, 94, 95, and 96, as well as the average of months (94, 95, and 96). You need to run a full range of fitted smoothing values (12 model runs) for the following:

Model Run #
1	.2	.2
2	.4	.2
3	.6	.2
4	.8	.2
5	.2	.4
6	.4	.4
7	.6	.4
8	.8	.4
9	.2	.6
10	.4	.6
11	.6	.6
12	.8	.6

13. as well as the optimal arima model,

1. Using Minitab 17.2 time series forecasting evaluation process (exponential smoothing model, and double exponential smoothing model) You need to forecast your insurance firm's stock price for months 94, 95, and 96 (October, November, and December 2015), based on your 93 months of data, fitted into your single exponential smoothing model and your double exponential smoothing. You need to compare your forecasted values to actual data for months 94, 95, and 96, (October, November, and December 2015), for each single exponential smoothing model, and each double exponential smoothing model. The results need to be summarized into a table comparing forecasting accuracy for all forecasting models simultaneously. You need to substantiate your choice of the best forecasting model for each of the three periods and over the total 3 month period (average forecast over the period October 2015-December 2015) with a thorough and complete discussion.