MGS 3100, week 4 notes
MGS 3100, week 4 notes MGS 3100
Popular in Buisness Analysis
Pysch 3 (Biological Basis of Psychology)
verified elite notetaker
verified elite notetaker
Popular in Managerial Science
This 4 page Class Notes was uploaded by Tricia Williams on Sunday July 3, 2016. The Class Notes belongs to MGS 3100 at Georgia State University taught by Mark Sweatt in Summer 2016. Since its upload, it has received 39 views. For similar materials see Buisness Analysis in Managerial Science at Georgia State University.
Reviews for MGS 3100, week 4 notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 07/03/16
MGS 3100 in-class notes (6/27/2016) Averaging techniques continue Simple exponential smoothing:- Looks at two things when forecasting, the actual sales and the forecast. The actual sales from the previous period are multiplied by the alpha, which is given, and one minus alpha multiplies the previous period forecast. The two amounts are then added together to give the forecast for the next period. Additionally, the second period forecast uses the naïve method of forecasting. Fitting a Trendline Regression: Tells if Y and X are related and if so, how closely related are they. Simple Regression:- Uses one variable to forecast the dependent variable Y ( Y=intercept + variable (coefficient)) Multiple Regression: Uses more than one variable or multiple independent variable to forecast the dependent variable Y. Y=Intercept + variable (coefficient) + variable (coefficient) + variable (coefficient)……………. This model is use when there is a trend in the data collected about the item to be forecasted. The above technique is evaluated by using the same evaluation criteria as averaging techniques plus R-squared. Regression equation= ( Y=intercept + variable (coefficient)) Mean shoe size = average of the shoe size Deviation from the mean = actual shoe size-mean shoe size SS= sum of squared deviations Degrees of freedom (df)- is the variance you are allow to have from forecasting P-value:- Tells which variable is most significant in relation to what is being forecast. (The smaller the p-value, the more significant). Significant of F:- How reliable is the model, the smaller the figure, the better. If it is not less than 0.1, then the correlation is not meaningful. Some truth about forecasting 1. What happen in the past is more likely to happen in the future. 2. Forecast accuracy increases for shorter time frames 3. Combined forecasting is more accurate than forecasting done on an individual basis. 4. Forecasts are hardly ever accurate. Formulas SS= df * MS 2 R-Square= SS/SST or (multiple R) Multiple R= Square root of R-Square Standard error= Square root of MS of the (residual) Total degrees of freedom (n-1) where n is the observation total Total df for regression = the number of variables=K Total df for residual = (n-K-1) Anova:- Analysis of Variance MGS 3100 in-class notes (6/29/2016) Trend and Seasonal Index/Seasonality All seasonality factors/indices must total to (4) Deseasonalized sales:- It is moving the seasonality from the data. (Actual sales/seasonalized index) Reseasonalized forecast (seasonalized) :- Making the final forecast. (deseasonalized forecast x seasonalized index (SI)) Forecasting: Decomposition of the trend and seasonality Moving average:- Take the average of the actual sales of two period above, the current period, and one period below. Centered average:- Take the average of the moving average current period and the immediate period that follow. Raw Indices: (Actual sales/centered average) Seasonalized index= averaging the raw indices for the different years (Year 1, year 2, year 3, year 4 etc.) Predicted sales (y-hat) are transferred from the output sheet. Y-Y hat error (Bias):- Sales (Y) - Predicted sales (Y-hat) 2 Error squared (Y-Y-hat) Finding the seasonalized forecast of a specific quarter Sales trend:- This is computed by first finding the time period for the specified quarter for which you need to find the seasonalized forecast (eg. Q1, 2013) Always begin with the beginning quarter as period one. You then plug the specified time period number into the sales trend equation, which will give the deseasonalized forecast. Seasonalized forecast: To find the seasonalized forecast is (seasonalized index for the specified quarter, in this case is Q1 multiplied by the deseasonalized forecast).
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'