Your data analysis solution
Name is required.
Email address is required.
Invalid email address
Answer is required.
Exceeding max length of 5KB

Back-transforming with multiple linear regression equation

By KRD | Oct 27, 2017 07:17PM CEST

I am running a multiple linear regression (MLR) and of the explanatory variables is log-transformed and the other is not. How do I use the MLR equation to predict the dependent variable? Specifically, do I use exponent in excel on the result? or exponent on just that variable that was log-transformed and the coefficient? For example, here 2.33 is a log-transformed value =EXP(2.47+((0.44*2.5)+(0.29*2.33))). Or does there need to be a EXP in front of the (0.29 *2.33) portion of the equation?

Best Answer
By Sébastien | Oct 30, 2017 09:48AM CET | XLSTAT Agent

Hi Kate,

For predictions, no need to enter any EXP. If the model equation was built over log-transformed variables, just enter the log transformed predictor values:

Y = 2.47 + 0.44*2.5 + 0.29*2.33

Jean Paul


Up

0

Down

By Sébastien | Oct 30, 2017 09:48AM CET | XLSTAT Agent

Hi Kate,

For predictions, no need to enter any EXP. If the model equation was built over log-transformed variables, just enter the log transformed predictor values:

Y = 2.47 + 0.44*2.5 + 0.29*2.33

Jean Paul

Post Your Public Answer

Contact Us

1c26995d494fb3061dd0ae8571ffc0a4@xlstat.desk-mail.com
https://cdn.desk.com/
false
desk
Loading
seconds ago
a minute ago
minutes ago
an hour ago
hours ago
a day ago
days ago
about
false
Invalid characters found
/customer/portal/articles/autocomplete
9283