Question: Forecasting Methods For Management Questions A client company (a Multistore Retailer) had experienced unprecedented volatility in the demand of some specific SKUs (stock keeping units)




A client company (a Multistore Retailer) had experienced unprecedented volatility in the demand of some specific SKUs (stock keeping units) from a single product category. The client desired to build a forecasting model to forecast the demand of certain SKUs, for 1 to 12 months into the future. A time-series dataset was constructed, including monthly data for the price, sales and approximately 50 current-period or lagged potential predictor variables. An autoregressive TimeSeries Model was developed to predict future demand. The client company used forecasted demand to better control costs of production \& inventory and to increase profitability. a. Assess the method and time horizon adopted by the client company. [10 Marks] b. Evaluate the potential problems on accuracy of forecasting faced by the company. [10 Marks] 2. TABLE Q2 below are the regression results of an SME. Details are as followS; a. Construct the model based on the findings above. [5 Marks] b. Interpret the findings by showing the signification of the variables and the model as a whole. [8 Marks] c. Discuss with some simulated figures on how the above regression model can contribute for prediction and forecasting. a. Differentiate between Autoregressive (AR) and Moving Average (MA) processes. [10 Marks] b. i. Assess the representation of p and q in the ARMA model. [5 Marks] ii. Explain the coefficients of Xti and ti in the equation below. You may give examples. Xt=c+i=1piXti+i=1qiti+t [5 Marks] a. Describe the importance of stationarity in time-series analysis. [8 Marks] b. Interpret the stationary and non-stationary graphs below. [12 Marks] 5. Time series analysis and forecasting are generally used in a highly volatile industry to analyze its trending sales over time. a. Propose any TWO (2) suitable methods in overcoming gross errors in forecasting by giving examples. [10 Marks] b. Assess the suitability of using multivariate modeling such as multiple regression in dealing with trends for forecasting purposes. [10 Marks] -END OF THE PAPER
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
