Scribd
Upload
79 views3 pages

Model Validation Measures: 1. R - Goodness of Fit Test

The document discusses model validation measures including R2, adjusted R2, standard error of estimate, and test statistics. It provides the mathematical expressions and interpretations for R2 and adjusted R2. It also discusses standard error of estimate and different hypothesis tests that can be performed using various test statistics.

Uploaded by

Shayan Sen Gupta
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
79 views3 pages

Model Validation Measures: 1. R - Goodness of Fit Test

The document discusses model validation measures including R2, adjusted R2, standard error of estimate, and test statistics. It provides the mathematical expressions and interpretations for R2 and adjusted R2. It also discusses standard error of estimate and different hypothesis tests that can be performed using various test statistics.

Uploaded by

Shayan Sen Gupta
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

Predictive Analytics

QM901x
Prof U Dinesh Kumar

Model Validation Measures

1. R2 – Goodness of Fit test

Measures Mathematical Expression Interpretation

Percentage of variation in Y
𝑆𝑆𝑅 𝑆𝑆𝐸 explained by the explanatory
R2 R2 = = 1-
𝑆𝑆𝑇 𝑆𝑆𝑇 variables in the model.

2 2 𝑛−1 Modification of R2(adjustment


Adjusted R Adjusted R = 1 − (1 − R2) 𝑛−𝑘−1
for degrees of freedom)

 Adjusted R2 is used to evaluate the value add to the model by


additional explanatory variables.

 Adjusted R2 < R2

 Adjusted R2 can be negative and may decrease when a new


explanatory variable is added to the model.

© All Rights Reserved, Indian Institute of Management Bangalore


Predictive Analytics
QM901x
Prof U Dinesh Kumar

2. Standard Error of Estimate

TERM EXPRESSION FOR STANDARD ERROR


(Estimate of Standard deviation)

ε (Error)


 ( Regression Coefficient )

Mean Value of Yi
(Mean of Y for a given Xi )

Value of Y for a given X

© All Rights Reserved, Indian Institute of Management Bangalore


Predictive Analytics
QM901x
Prof U Dinesh Kumar

3. Test Statistic:

Hypothesis Test Statistic

H0 : β = 0

H1: β ≠ 0

H0 : β = 0 All Betas are zero

H1: β ≠ 0 Not all Betas are zero

H0 : β = 0 All newly added betas are


zero

H1: β ≠ 0 Not all newly added betas


are zero

© All Rights Reserved, Indian Institute of Management Bangalore

You might also like