Scribd
Upload
103 views7 pages

Regression Analysis Using Excel: X Abp

This document discusses using regression analysis in Excel to fit linear demand curves to data points. It explains how regression analysis finds the parameter values (a and b) that minimize the sum of squared errors between the data points and the linear curve. The example shows using Excel's Data Analysis tool to perform a regression on sample demand and price data, which estimates the demand curve as x1 = 433 - 142p1. It then performs multiple regression to include income and another price variable, estimating the demand curve as x1 = -228.2 + 1.5I - 38.2p1 - 12.9p2.

Uploaded by

Bonaventure Nzeyimana
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
103 views7 pages

Regression Analysis Using Excel: X Abp

This document discusses using regression analysis in Excel to fit linear demand curves to data points. It explains how regression analysis finds the parameter values (a and b) that minimize the sum of squared errors between the data points and the linear curve. The example shows using Excel's Data Analysis tool to perform a regression on sample demand and price data, which estimates the demand curve as x1 = 433 - 142p1. It then performs multiple regression to include income and another price variable, estimating the demand curve as x1 = -228.2 + 1.5I - 38.2p1 - 12.9p2.

Uploaded by

Bonaventure Nzeyimana
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

Regression Analysis using Excel1 

Simple regression

Use Solver and some simple utility function to compute some demands for commodity 1 as a
function of income and prices. Here is a small sample

We now try to fit a linear demand curve

x1 = a − bp1
The data points are depicted below.

                                                            
1
 These notes are almost identical to those in the Regression Analysis Slides 
Clearly no line is going to pass through each point. Regression analysis starts with an initial
guess as to the values of the parameters a and b. It then computes the vertical distance between
the line and each dot and then sums the square of these distances.
The regression program then finds the parameters that minimize this sum of “squared errors.”
The estimated parameter vector is then called the least squares estimate.

From the Home Ribbon, click on Data. On the far right you will see the Analysis Options

Click on Data Analysis and scroll down to Regression.

Click on OK.
You need to enter the Y (dependent variable data). This is the array of demands [D3:D8].

You also need to enter the X (independent variable array.) This is the array [B3:B8].

You do want a constant term so leave the “constant is zero” box blank.

Note that the default is for the regression results to appear on a new worksheet.

If would like to see them on the same worksheet you need to tell the program where to write the
results. Om my spreadsheet everything is blank to the South-West of A10 so I choose the region
A10:P30.
Click OK and the results will appear.

SUMMARY OUTPUT

We are interested in the coefficients. What the SUMMARY OUTPUT reveals is that the least
squares estimate is

x1 = 433 −142 p1
The R2 is 0.85 indicating that we have “explained” 85% of the variation.

 
 

Drawing the chart 

Using the estimated coefficients we can compute the estimated demand for each price. 

 
 

We can then plot the estimated demand curve on the same chart as the actual scatter plot. 

 
 

 
 
Multiple Regression 

Simple regression leaves out the fact that the variation in  x1  results not just from the variation 
in  p1  but in the other price and income as well.  The analysis proceeds as before but now the X‐
array is  J3:L8.  The new SUMMARY OUTPUT IS 

 
 

Thus the least squares estimate is    x1 = −228.2 + 1.5* I − 38.2* p1 −12.9* p2 . 

We should not put too much faith in this result as there are only 6 observations and 4 
explanatory variables (including the constant.)               
          

You might also like