The following table lists the worldwide shipments of personal computers (in thousands) according to Dataquest.
Year Shipments (in thousands)
(a) Use exponential smoothing to determine the forecast of shipments for the year 1999. Use the actual shipments for 1990 as the starting forecast for 1991. Use a smoothing constant of ? = 0.4.
(b) Plot the data, fit a trend line, display equation and R2 on the chart and discuss the strength of prediction of the regression model. Note: Use 1990 = 1, 1991 = 2, and so on to get more accurate equation.
(c) Use the regression model to predict the shipments for the years 1992 through 1999.
(d) Which forecasting method would you prefer to use and why? Note: Compute MAPE for the two forecasting methods to compare the accuracy for the period 1992-1998.
I also did step of (a) and those values are in column G. Please check if it is consistent with yours. For (b), I plotted the trend line in the excel ...
In this solution, an example for exponential smoothing for forecasting is provided. In addition, comparison between exponential smoothing and linear regression is made based on MAPE.
Questions: Moving Averages, Wine Sales, Tax Prediction, Income, etc.
1) The method of moving averages is used:
(A) to plot a series
(B) to exponentiate a series
(C) to smooth a series
(D) in regression analysis
2) The following table contains the number of complaints received in a department store for the first 6 months of last year.
Referring to the above table, If a three-term moving average is used to smooth this series, what would be the second calculated term?
3) In multiple regression, the _____ procedure permits variables to enter and leave the model at different stages of its development
(A) forward selection
(B) residual analysis
(C) backward elimination
(D) stepwise regression
4) A real estate builder wishes to determine how house size (House) is influenced by the family income(Income ) family size (SIZE) , and education of the head of the household (School). House size is measured in hundreds of square feet, income is measured in thousand of dollars and education is measure in years. The builders randomly selected 50 families and ran the multiple regressions The business literature involving human capital shows the education influence an Individual's annual income. Combined, these may influence family size. With this in mind, what should the real estate builder be particularly concern with when analyzing the multiple regression models.
(A) randomness of error terms
(C) normality of residuals
(D) missing observation
5) An auditor for a country government would like to develop a model to predict the county taxes based on the age of single family houses. A random sample of single-family houses has been selected, with the results as shown below:
Taxes Age of house
Assuming a quadratic relationship between the age of the houses and the country taxes, which of the following is the best prediction of the average county taxes for a 20 year old house
(A) $ 557.30
6) When using the exponentially weighted moving average for purpose of forecasting rather than smoothing
(A) the previous smoothed value become the forecast
(B) the current smoothed value becomes the forecast
(C) the next smoothed value become the forecast
(D) none of the above
7)You need to decide whether you should invest in a particular stock. You would like to invest if the price is likely to rise in the long run. You have data on the daily average price of this stock over the last 12 months . Your best action is to
(A) compute moving average
(B) perform exponential smoothing
(C) estimate a least square trend model
(D) compute the MD statistic
8) The number of cases of merlot wine sold by a Paso Robles winery in an 8 year period follows
Year Cases of wine
Referring to the table above, the holt-winter method for forecasting with smoothing constant of 0.2 for both level and trend will be used to smooth the wine sale. The smoothed value of the level and trend for 1993 are _____ and _____ respectively.
(A) 356 and 86
(B) 406.8 and 57.84
(C) 398 and 61.25
(D) 374.8 and 76.5
9) A model that can be used to make prediction about long-term future valves of ta time series is:
(A) Linear Trend
(B) quadratics trend
(C) exponential trend
(D) all of the above