Otdfbt

What word describes the sound of an instrument based on the shape of the waveform of its sound?

What is more safe for browsing the web: PC or smartphone?

Why would a military not separate its forces into different branches?

Referring to person by surname, keep or omit "von"?

Where to draw the line between quantum mechanics theory and its interpretation(s)?

As a GM, is it bad form to ask for a moment to think when improvising?

How can I obtain and work with a Platonic dodecahedron?

How long does it take a postcard to get from USA to Germany?

If I no longer control a permanent when it would be sacrificed, is it still sacrificed?

Why is the blank symbol not considered part of the input alphabet of a Turing machine?

How would you join two twin beds?

All of my Firefox add-ons been disabled suddenly, how can I re-enable them?

Two denim hijabs

How long did it take Captain Marvel to travel to Earth?

How to speed up large double sums in a table?

Can a player choose to add detail and flavor to their character's spells and abilities?

Picking a theme as a discovery writer

How is Pauli's exclusion principle still valid in these cases?

My large rocket is still flipping over

Is there any other simpler way to draw the following cross section?

Does Thanos's ship land in the middle of the battlefield in "Avengers: Endgame"?

Given a safe domain, are subdirectories safe as well?

What detail can Hubble see on Mars?

Copper as an adjective to refer to something made of copper

Interpret a multiple linear regression when Y is log transformed [duplicate]

Interpretation of log transformed predictor and/or responseHow to interpret logarithmically transformed coefficients in linear regression?Does the first line of interaction output from lmer contain all levels of variables?Interpretation of log transformed coefficients, OLS regressionDummy variables in multiple regressionCoefficients linear and log-linear regression modelHow to find y values given x values and summary statistics on yInteraction effects with three dummy variables - interpretationInterpretation of variable in a probit modelinterpreting HLM coefficientsInterpretation of interaction term in a probit estimation

.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty margin-bottom:0;

4

$begingroup$

This question already has an answer here:

• 
  Interpretation of log transformed predictor and/or response
  
  3 answers
  
  

I have the following multiple linear regression model:
Log(y) = B0 + B1X1 + B2X2 + B3x3 + e.
X1 is a dummy that can take 0 = male and 1 = female and
X2 and X3 are continuous variables.

I am not entirely sure on how to interpret the coefficients for the variables.
The coefficient for the dummy variable is 0,20. Does that mean, that changing from male to female (male is baseline) the Y will increase by an average of 20%. Is it directly translated into percentage?

And for the continuous variables, the coefficient for X2 is 0,1. Does that mean that increasing X2 with 1 unit increases Y with an average of 10%? Again, is it directly translated into percentage?

regression data-transformation interpretation logarithm

share|cite|improve this question

edited Apr 27 at 17:42

Richard Hardy

28.6k645132

asked Apr 27 at 14:37

maSmaS

491

$endgroup$

marked as duplicate by COOLSerdash, Richard Hardy, Siong Thye Goh, kjetil b halvorsen, Michael Chernick Apr 27 at 20:49

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

add a comment | 

4

$begingroup$

This question already has an answer here:

• 
  Interpretation of log transformed predictor and/or response
  
  3 answers
  
  

I have the following multiple linear regression model:
Log(y) = B0 + B1X1 + B2X2 + B3x3 + e.
X1 is a dummy that can take 0 = male and 1 = female and
X2 and X3 are continuous variables.

I am not entirely sure on how to interpret the coefficients for the variables.
The coefficient for the dummy variable is 0,20. Does that mean, that changing from male to female (male is baseline) the Y will increase by an average of 20%. Is it directly translated into percentage?

And for the continuous variables, the coefficient for X2 is 0,1. Does that mean that increasing X2 with 1 unit increases Y with an average of 10%? Again, is it directly translated into percentage?

regression data-transformation interpretation logarithm

share|cite|improve this question

edited Apr 27 at 17:42

Richard Hardy

28.6k645132

asked Apr 27 at 14:37

maSmaS

491

$endgroup$

marked as duplicate by COOLSerdash, Richard Hardy, Siong Thye Goh, kjetil b halvorsen, Michael Chernick Apr 27 at 20:49

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

add a comment | 

$begingroup$

This question already has an answer here:

• 
  Interpretation of log transformed predictor and/or response
  
  3 answers
  
  

I have the following multiple linear regression model:
Log(y) = B0 + B1X1 + B2X2 + B3x3 + e.
X1 is a dummy that can take 0 = male and 1 = female and
X2 and X3 are continuous variables.

I am not entirely sure on how to interpret the coefficients for the variables.
The coefficient for the dummy variable is 0,20. Does that mean, that changing from male to female (male is baseline) the Y will increase by an average of 20%. Is it directly translated into percentage?

And for the continuous variables, the coefficient for X2 is 0,1. Does that mean that increasing X2 with 1 unit increases Y with an average of 10%? Again, is it directly translated into percentage?

regression data-transformation interpretation logarithm

share|cite|improve this question

edited Apr 27 at 17:42

Richard Hardy

28.6k645132

asked Apr 27 at 14:37

maSmaS

491

$endgroup$

share|cite|improve this question

edited Apr 27 at 17:42

Richard Hardy

28.6k645132

asked Apr 27 at 14:37

maSmaS

491

share|cite|improve this question

edited Apr 27 at 17:42

Richard Hardy

28.6k645132

asked Apr 27 at 14:37

maSmaS

491

edited Apr 27 at 17:42

Richard Hardy

28.6k645132

edited Apr 27 at 17:42

Richard Hardy

28.6k645132

asked Apr 27 at 14:37

maSmaS

491

asked Apr 27 at 14:37

maSmaS

491

1 Answer
1

active

oldest

votes

4

$begingroup$

Positive coefficients somehow indicate a positive effect, but they don't simply turn into percentages. There is a transformation. Let's say your model is $log y = b_0+b_1x_1$; this means $y=e^b_0+b_1x_1=A_0e^b_1x_1$. So, dummy or not, if $x_1$ increases by $1$ unit, $y$ increases by $e^b_1$, i.e. if $b_1=0.2$, $y$ increases by $e^0.2approx 1.22$, i.e. $22%$. The case is similar for your continuous variable.

share|cite|improve this answer

answered Apr 27 at 15:20

gunesgunes

8,8441419

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  I have read that if the natural log is used it is approximately translated into a percentage change, if the change in x is small. Is that correct?
  $endgroup$
  – maS
  Apr 27 at 15:45
  

  
  
  
  


• 
  
  
  

  
  1
  

  
  
  
  
  
  

  
  $begingroup$
  Yes, but approximate is the key term here, because Taylor expansion of $e^x$ is: $e^x=1+x+x^2/2!,...approx 1+x$ when $x$ is small. Which means $x%$ increase in something. Here, to execute this idea, your coefficients should be small.
  $endgroup$
  – gunes
  Apr 27 at 15:47
  
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  Thanks gunes. Would you recommend using natural log then as my log transformation or what base should i use? cant seem to find a solid explanation of the choice
  $endgroup$
  – maS
  Apr 27 at 16:08
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  Also: inference on transformed variables $ne$ inference on un-transformed variables.
  $endgroup$
  – Alexis
  Apr 27 at 16:57
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  @maS using base N results in model $y=e^b_0log N+b_1log N x_1$, where you solve for $a_i=b_ilog N$. It's one to one.
  $endgroup$
  – gunes
  Apr 27 at 17:36
  

  
  
  
  

add a comment | 

4

$begingroup$

Positive coefficients somehow indicate a positive effect, but they don't simply turn into percentages. There is a transformation. Let's say your model is $log y = b_0+b_1x_1$; this means $y=e^b_0+b_1x_1=A_0e^b_1x_1$. So, dummy or not, if $x_1$ increases by $1$ unit, $y$ increases by $e^b_1$, i.e. if $b_1=0.2$, $y$ increases by $e^0.2approx 1.22$, i.e. $22%$. The case is similar for your continuous variable.

share|cite|improve this answer

answered Apr 27 at 15:20

gunesgunes

8,8441419

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  I have read that if the natural log is used it is approximately translated into a percentage change, if the change in x is small. Is that correct?
  $endgroup$
  – maS
  Apr 27 at 15:45
  

  
  
  
  


• 
  
  
  

  
  1
  

  
  
  
  
  
  

  
  $begingroup$
  Yes, but approximate is the key term here, because Taylor expansion of $e^x$ is: $e^x=1+x+x^2/2!,...approx 1+x$ when $x$ is small. Which means $x%$ increase in something. Here, to execute this idea, your coefficients should be small.
  $endgroup$
  – gunes
  Apr 27 at 15:47
  
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  Thanks gunes. Would you recommend using natural log then as my log transformation or what base should i use? cant seem to find a solid explanation of the choice
  $endgroup$
  – maS
  Apr 27 at 16:08
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  Also: inference on transformed variables $ne$ inference on un-transformed variables.
  $endgroup$
  – Alexis
  Apr 27 at 16:57
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  @maS using base N results in model $y=e^b_0log N+b_1log N x_1$, where you solve for $a_i=b_ilog N$. It's one to one.
  $endgroup$
  – gunes
  Apr 27 at 17:36
  

  
  
  
  

add a comment | 

4

$begingroup$

Positive coefficients somehow indicate a positive effect, but they don't simply turn into percentages. There is a transformation. Let's say your model is $log y = b_0+b_1x_1$; this means $y=e^b_0+b_1x_1=A_0e^b_1x_1$. So, dummy or not, if $x_1$ increases by $1$ unit, $y$ increases by $e^b_1$, i.e. if $b_1=0.2$, $y$ increases by $e^0.2approx 1.22$, i.e. $22%$. The case is similar for your continuous variable.

share|cite|improve this answer

answered Apr 27 at 15:20

gunesgunes

8,8441419

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  I have read that if the natural log is used it is approximately translated into a percentage change, if the change in x is small. Is that correct?
  $endgroup$
  – maS
  Apr 27 at 15:45
  

  
  
  
  


• 
  
  
  

  
  1
  

  
  
  
  
  
  

  
  $begingroup$
  Yes, but approximate is the key term here, because Taylor expansion of $e^x$ is: $e^x=1+x+x^2/2!,...approx 1+x$ when $x$ is small. Which means $x%$ increase in something. Here, to execute this idea, your coefficients should be small.
  $endgroup$
  – gunes
  Apr 27 at 15:47
  
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  Thanks gunes. Would you recommend using natural log then as my log transformation or what base should i use? cant seem to find a solid explanation of the choice
  $endgroup$
  – maS
  Apr 27 at 16:08
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  Also: inference on transformed variables $ne$ inference on un-transformed variables.
  $endgroup$
  – Alexis
  Apr 27 at 16:57
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  @maS using base N results in model $y=e^b_0log N+b_1log N x_1$, where you solve for $a_i=b_ilog N$. It's one to one.
  $endgroup$
  – gunes
  Apr 27 at 17:36
  

  
  
  
  

add a comment | 

$begingroup$

Positive coefficients somehow indicate a positive effect, but they don't simply turn into percentages. There is a transformation. Let's say your model is $log y = b_0+b_1x_1$; this means $y=e^b_0+b_1x_1=A_0e^b_1x_1$. So, dummy or not, if $x_1$ increases by $1$ unit, $y$ increases by $e^b_1$, i.e. if $b_1=0.2$, $y$ increases by $e^0.2approx 1.22$, i.e. $22%$. The case is similar for your continuous variable.

share|cite|improve this answer

answered Apr 27 at 15:20

gunesgunes

8,8441419

$endgroup$

share|cite|improve this answer

answered Apr 27 at 15:20

gunesgunes

8,8441419

answered Apr 27 at 15:20

gunesgunes

8,8441419

answered Apr 27 at 15:20

gunesgunes

8,8441419