Scribd
Upload
5 views

Lecture 1

The document provides an introduction to linear regression, focusing on its application in financial analysis to explain variations in dependent variables, such as return on assets (ROA), using independent variables like capital expenditures (CAPEX). It outlines the process of estimating regression lines, interpreting coefficients, and the assumptions underlying simple linear regression, including linearity and normality of residuals. Additionally, it discusses measures of goodness of fit, such as the coefficient of determination and F-statistic, to assess the statistical significance of the regression model.

Uploaded by

amir rafique
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
5 views

Lecture 1

The document provides an introduction to linear regression, focusing on its application in financial analysis to explain variations in dependent variables, such as return on assets (ROA), using independent variables like capital expenditures (CAPEX). It outlines the process of estimating regression lines, interpreting coefficients, and the assumptions underlying simple linear regression, including linearity and normality of residuals. Additionally, it discusses measures of goodness of fit, such as the coefficient of determination and F-statistic, to assess the statistical significance of the regression model.

Uploaded by

amir rafique
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 38

Introduction to Linear

Regression
• Financial analysts often need to examine whether a variable is
useful for explaining another variable.
• For example, the analyst may want to know whether earnings
growth, or perhaps cash flow growth, helps explain the
company’s value in the marketplace.
• Regression analysis is a tool for examining this type of issue
• Suppose an analyst is examining the return on assets (ROA) for
an industry and observes the ROA for the six companies shown
in Exhibit 1.
• The average of these ROAs is 12.5%, but the range is from 4%
to 20%.
• In trying to understand why the ROAs differ among these companies, we could
look at why the ROA of Company A differ from that of Company B, why the
ROA of Company A differ from that of Company D, why the ROA of Company
F differ from that of Company C, and so on, comparing each pair of ROAs.
• A way to make this a simpler exploration is to try to understand why each
company’s ROA differ from the mean ROA of 12.5%.
• We look at the sum of squared deviations of the observations
from the mean to capture variations in ROA from their mean.
• Let Y represent the variable that we would like to explain,
which in this case is the return on assets. Let Yi represent an
observation of a company’s ROA, and let represent the mean
ROA for the sample of size n. We can describe the variation of
the ROAs as
• Our goal is to understand what drives these returns on assets or, in other
words, what explains the variation of Y. The variation of Y is often referred to
as the sum of squares total (SST), or the total sum of squares.
• We now ask whether it is possible to explain the variation of the ROA using
another variable that also varies among the companies; note that if this
other variable is constant or random, it would not serve to explain why the
ROAs differ from one another.
• Suppose the analyst believes that the capital expenditures in the previous
period, scaled by the prior period’s beginning property, plant, and
equipment, are a driver for the ROA variable.
• Let us represent this scaled capital expenditures variable as CAPEX, as we
show in Exhibit 2.
†e The variation of X, in this case CAPEX, is calculated as

We can see the relation between ROA and CAPEX in the scatter
plot
• We refer to the variable whose variation is being explained as the dependent
variable, or the explained variable; it is typically denoted by Y.
• We refer to the variable whose variation is being used to explain the variation
of the dependent variable as the independent variable, or the explanatory
variable; it is typically denoted by X.
• In our example, the ROA is the dependent variable (Y) and CAPEX is the
independent variable (X).
• A common method for relating the dependent and independent variables is
through the estimation of a linear relationship, which implies describing the
relation between the two variables as represented by a straight line.
• If we have only one independent variable, we refer to the method as simple
linear regression (SLR);
• if we have more than one independent variable, we refer to the method as
multiple regression.
• Linear regression allows us to test hypotheses about the relationship between
two variables, by quantifying the strength of the relationship between the two
variables, and to use one variable to make predictions about the other
variable.
• Our focus is on linear regression with a single independent variable—that is,
simple linear regression.
Identifying the Dependent and Independent Variables in a
Regression

• An analyst is researching the relationship between corporate


earnings growth and stock returns. Specifically, she is
interested in whether earnings revisions affect stock price
returns in the same period. She collects five years of monthly
data on “Wall Street” EPS revisions for a sample of 100
companies and on their monthly stock price returns over the
five year period.
The Basics of Simple Linear Regression

Intercept,
Slope
coefficient
Error term
Estimating the Regression Line
• Fitting the line requires minimizing the sum of the squared
residuals, the sum of squares error (SSE), also known as the
residual sum of squares:
Once we estimate the slope, we can then estimate the intercept
using the mean of Y and the mean of X:
Interpreting the Regression Coefficients

• The return on assets for a company is 4.875% if the company


makes no capital expenditures.

• If CAPEX increases by one unit—say, from 4% to 5%—ROA


increases by 1.25%.
Assumptions of the simple linear
regression
1. Linearity: The relationship between the dependent variable, Y,
and the independent variable, X, is linear.
2. Homoskedasticity: The variance of the regression residuals is
the same for all observations.
3. Independence: The observations, pairs of Ys and Xs, are
independent of one another. This implies the regression
residuals are uncorrelated across observations.
4. Normality: The regression residuals are normally distributed.
Linearity
Linearity
Homoskedasticity
Independence
Normality
• Normality applies to the residuals NOT individual
independent and dependent variables
• Central Limit Theorem relax this assumption for large
samples
Breakdown of Variation of
Dependent Variable
Measures of Goodness of Fit
• Coefficient of determination, the F-statistic for the test of fit,
and the standard error of the regression
F Statistic
• Whereas the coefficient of determination—the portion of the
variation of the dependent variable explained by the
independent variable—is descriptive, it is not a statistical test.
• To see if our regression model is likely to be statistically
meaningful, we will need to construct an F-distributed test
statistic.
Standard error of
estimate/regression
Functional forms for simple linear
regression
Log-lin model
Lin-Log Model
Log-Log Model
Selecting the correct functional form
• The key to fitting the appropriate functional form of a simple
linear regression is examining the goodness of fit measures—
• The coefficient of determination (R2),
• The F-statistic, and
• The standard error of the estimate
• As well as examining whether there are patterns in the
residuals.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy