Projection and Regression | Regression Analysis | Errors And Residuals

# 10/17/2008

9152034.xls

Projection & Regression in Excel
Excel offers a number of means to project future values and to perform regression analysis: Use of the Regression tool in the Analysis Toolpak add-in (TookPak) Adding a trendline to a chart (Chart Trendline) Completing values with the Series command (Series) Projecting values with the FORECAST function (Functions) Projecting values with the TREND or GROWTH functions (Functions) Calculating a straight line or exponential curve from existing data using the LINEST or LOGEST functions (Functions)

The Regression Tool in the ToolPak Add-In
Finding the Regression Tool Part of Excel is an add-in named the "Analysis ToolPak". When you choose the Tools menu, if you don't see the option "Data Analysis…" in the list, you need to add in or install and add in the Analysis ToolPak. Starting the Regression Tool When you choose Tools, Data Analysis from the menus, a dialog like the one below displays:

Regression is one of the analysis tools in the Analysis ToolPak. Completing the Regression Dialog Excel displays a "Regression" dialog that prompts for information about your data and about what results you want to see. Complete the dialog and click "OK". A Small Example quantity ads 27500 10 25000 7.5 25000 10 25000 5 27500 10 22500 7.5 25000 7.5 22500 5 27500 5 25000 5 27500 10 25000 5 22500 5 22500 5

Excel's Regression Output The look of your results will depend on what output options you selected in the dialog and on how much data you have. You may need to widen columns in order to see the output completely. SUMMARY OUTPUT Regression Statistics Multiple R 0.55 R Square 0.3 Adjusted R Square 0.24 Standard Error 1.94 Observations 14 ANOVA df Regression Residual Total Intercept quantity 1 12 13 SS 19.53 45.2 64.73 MS 19.53 3.77 F Significance F 5.19 0.04

Coefficients Standard Error t Stat -8.66 6.88 -1.26 0 0 2.28

P-value Lower 95%Upper 95% Lower 95.0% 0.23 -23.65 6.33 -23.65 0.04 0 0 0

Notes For the Y Range Enter the reference for the range of dependent data. The range must consist of a single column of data. For the X Range Enter the reference for the range of independent data. Excel orders independent variables from this range in ascending order from left to right. The maximum number of independent variables is 16. The data used in the regression must be contiguous in the worksheet.

From Excel's menus select Tools, Add-ins and toggle on the checkmark to the left of "Analysis ToolPak".

Upper 95.0% 6.33 0

Trendlines in Charts
You can extend a trendline in a chart beyond the actual data to predict future values. How To To add a trendline to a data series in a chart: 1. Select the data series to which you want to add a trendline. 2. From the menus choose Chart, Add Trendline. 3. On the Type tab, click the type of regression trendline or moving average you want. 4. On the Options tab, choose how many periods forward or back you want to forecast. If you select Polynomial, enter in the Order box the highest power for the independent variable. If you select Moving Average, enter in the Period box the number of periods to be used to calculate the moving average. A Small Example Quarterly Revenues Q1 2 Q2 2.2 Q3 2.5 Q4 2.55 Chart of Actual Data
Revenue in Millions 2.6 2.5 2.4 2.3 2.2 2.1 2 1.9 1.8

Q1

Q2

Q3

Q4

Chart w/4 Quarter Forecast
Revenue in Millions 2.6 2.5 2.4 2.3 2.2 2.1 2 1.9 1.8

Q1

Q2

Q3

Q4

The equation and R-squared value are displayed on the chart.

Using the Series Command
Use the Edit, Fill, Series commands to project future values and perform regression analysis. A Linear Example Initial 1 2

Projected 1 2 3 4 5

A Growth Example Initial 1 3

Projected 1 1.5 2.25 3.38 5.06

How To Select the initial data and as many rows beneath it as you want forecast values. Choose Edit, Fill, Series and complete the "Series" dialog. Click OK.

Projection Functions
FORECAST

TREND

GROWTH

LINEST

LOGEST