Help the Stat Consulting Group by giving a gift

How do I interpret a regression model when some variables are log transformed?

In this page, we will discuss how to interpret a regression model when some
variables in the model have been log transformed.
The example data can be downloaded here (the file is in .csv format). The
variables in the data set are writing, reading, and math scores (**write**, **read** and **math)**,
the log transformed writing (**lgwrite**) and log transformed math scores (**lgmath**)
and **female**. For these examples, we have taken the natural log (ln). All the examples are done in Stata, but they can be easily
generated in any statistical package. In the examples below, the variable **write** or its log
transformed version will be used as the outcome variable. The examples are used for
illustrative purposes and are not intended to make substantive sense. Here is a
table of different types of means for variable **write**.

Variable | Type Obs Mean [95% Conf. Interval] -------------+---------------------------------------------------------- write | Arithmetic 200 52.775 51.45332 54.09668 | Geometric 200 51.8496 50.46854 53.26845 | Harmonic 200 50.84403 49.40262 52.37208 ------------------------------------------------------------------------

Very often, a linear relationship is hypothesized between a log transformed outcome variable and a group of predictor variables. Written mathematically, the relationship follows the equation

log(y_i)= β

_{0 }+ β_{1}*x1 + ... + β_{k}*xk + e_i

where y is the outcome variable and x1, .., xk are the predictor variables.
In other words, we assume that log(y) - **x**'**β**
is normally distributed, (or y is log-normal conditional on all the covariates.) Since this is just
an ordinary least squares regression, we can easily interpret a
regression coefficient, say β_{1},_{
}as the expected change in log of y with respect to a one-unit
increase in x1 holding all other variables at any fixed value, assuming that x1
enters the model only as a main effect. But what if we want to know
what happens to the outcome variable **y** itself for a one-unit increase in x1?
The natural way to do this is to interpret the exponentiated regression
coefficients, exp(β), since exponentiation is
the inverse of logarithm function.

Let's start with the intercept-only model, log(**write**) = β_{0}.

------------------------------------------------------------------------------ lgwrite | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- intercept | 3.948347 .0136905 288.40 0.000 3.92135 3.975344 ------------------------------------------------------------------------------

We can say that 3.95 is the unconditional expected mean of log of write.
Therefore the exponentiated value is exp(3.948347) = 51.85. This is the geometric mean of
**write**. The emphasis here is that it is the geometric mean
instead of the arithmetic mean. OLS regression of the original variable **y** is used to
to estimate the expected arithmetic mean and OLS regression of the log
transformed outcome variable is to estimated the expected geometric mean of the
original variable.

Now let's move on to a model with a single binary predictor variable.

------------------------------------------------------------------------------ lgwrite | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- female | .1032614 .0265669 3.89 0.000 .050871 .1556518 intercept | 3.89207 .0196128 198.45 0.000 3.853393 3.930747 ------------------------------------------------------------------------------

log(write)= β

_{0 }+ β_{1}*female= 3.89 + .10*female

Before diving into the interpretation of these parameters, let's get the
means of our dependent variable, **write**, by gender.

males Variable | Type Obs Mean [95% Conf. Interval] -------------+---------------------------------------------------------- write | Arithmetic 91 50.12088 47.97473 52.26703 | Geometric 91 49.01222 46.8497 51.27457 | Harmonic 91 47.85388 45.6903 50.23255 ------------------------------------------------------------------------ females Variable | Type Obs Mean [95% Conf. Interval] -------------+---------------------------------------------------------- write | Arithmetic 109 54.99083 53.44658 56.53507 | Geometric 109 54.34383 52.73513 56.0016 | Harmonic 109 53.64236 51.96389 55.43289 ------------------------------------------------------------------------

Now we can map the parameter estimates to the geometric means for the two
groups. The intercept of 3.89 is the log of geometric mean of **write** when
female = 0, i.e., for males. Therefore, the exponentiated value of it is the
geometric mean for the male group: exp(3.892) = 49.01. What can we say about the
coefficient for **female**? In the log scale, it is the difference in
the expected geometric means of the log of **write** between the female
students and male students. In the original scale of the variable **write**,
it is the ratio of the geometric mean of **write** for female students over
the geometric mean of **write** for male students, exp(.1032614) =
54.34383/49.01222 = 1.11. In terms of percent change, we can say that switching
from male students to female students, we expect to see about 11% increase in
the geometric mean of writing scores.

Last, let's look at a model with multiple predictor variables.

------------------------------------------------------------------------------ lgwrite | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- female | .114718 .0195341 5.87 0.000 .076194 .153242 read | .0066305 .0012689 5.23 0.000 .0041281 .0091329 math | .0076792 .0013873 5.54 0.000 .0049432 .0104152 intercept | 3.135243 .0598109 52.42 0.000 3.017287 3.253198 ------------------------------------------------------------------------------log(

write)= β_{0 }+ β_{1}*female+ β_{2}*read+ β_{3}*math

The exponentiated coefficient exp(β_{1})
for **female** is the ratio of the expected geometric mean for the female
students group over the expected geometric mean for the male students group, when
**read** and **math** are held at some fixed value. Of course, the
expected geometric means for the male and female students group will be
different for different values of **read** and **math**. However, their ratio is a
constant: exp(β_{1}). In our
example, exp(β_{1}) = exp(.114718)
= 1.12. We can say that writing scores will be 12% higher for the female
students than for the male students. For the variable **read**, we
can say that for a one-unit increase in **read**, we expect to see about a 0.7%
increase in writing score, since exp(.0066305) = 1.006653.
For a ten-unit increase in **read**, we expect to see about a 6.9% increase in writing score, since exp(.0066305*10) = 1.0685526.
The intercept becomes less interesting when the predictor variables are not centered and are continuous. In this particular model, the intercept is the expected
mean for log(write) for male (**female** =0) when read and math are equal to zero.

In summary, when the outcome variable is log transformed, it is natural to interpret the exponentiated regression coefficients. These values correspond to changes in the ratio of the expected geometric means of the original outcome variable.

Occasionally, we also have some predictor variables being log transformed. In this section, we will take a look at an example where some predictor variables are log-transformed, but the outcome variable is in its original scale.

------------------------------------------------------------------------------ write | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- female | 5.388777 .9307948 5.79 0.000 3.553118 7.224436 lgmath | 20.94097 3.430907 6.10 0.000 14.17473 27.7072 lgread | 16.85218 3.063376 5.50 0.000 10.81076 22.89359 intercept | -99.16397 10.80406 -9.18 0.000 -120.4711 -77.85685 ------------------------------------------------------------------------------

Written in equation, we have

write= β_{0 }+ β_{1}*female+ β_{2}*lgmath+ β_{3}*lgread

Since this is an OLS regression, the interpretation of the regression
coefficients for the non-transformed variables are unchanged from an OLS
regression without any transformed variables. For example, the
expected mean difference in writing scores between the female and male students is about 5.4 points, holding
the other predictor variables constant. On the other hand,
due to the log transformation, the estimated effects of **math** and **read** are no
longer linear, even though the effect of **lgmath** and **lgread** are
linear. The plot below shows the curve of predicted values against the reading
scores for the female students group holding math score constant.

How do we interpret the coefficient of 16.85218 for the variable of log of
reading score? Let's take two values of reading score, r1 and r2. The expected
mean difference in writing score at r1 and r2, holding the other predictor variables constant,
is **write**(r2) - **write**(r1) =
β_{3}*(log(r2) - log(r1)) = β_{3}*log(r2/r1).
This means that as long as the percent increase in **read**
(the predictor variable) is fixed, we will see the same difference in writing
score, regardless where the baseline reading score is. For example, we can say
that for a 10% increase in reading score, the difference in the expected mean
writing scores will be always β_{3}*log(1.10)
= 16.85218*log(1.1) = 1.61.

What happens when both the outcome variable and predictor variables are log transformed? We can combine the two previously described situations into one. Here is an example of such a model.

------------------------------------------------------------------------------ lgwrite | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- female | .1142399 .0194712 5.87 0.000 .07584 .1526399 lgmath | .4085369 .0720791 5.67 0.000 .2663866 .5506872 read | .0066086 .0012561 5.26 0.000 .0041313 .0090859 intercept | 1.928101 .2469391 7.81 0.000 1.441102 2.415099 ------------------------------------------------------------------------------

Written as an equation, we can describe the model:

log(**write**)= β_{0 }+ β_{1}***female**
+ β_{2}*log**(math)** + β_{3}***read**

For variables that are not transformed, such as **female**, its
exponentiated coefficient is the ratio of the geometric
mean for the female to the geometric mean for the male students group. For
example, in our example, we can say that the expected percent increase in
geometric mean from male student group to female student group is about 12%
holding other variables constant, since exp(.1142399) = 1.12. For reading score,
we can say that for a one-unit increase in reading score, we expected to see
about 0.7% of increase in the geometric mean of writing score, since exp(.0066086)
= 1.007.

Now, let's focus on the effect of **math**. Take two values of **math**, m1
and m2, and hold the other predictor variables at any fixed value. The equation above yields

log(**write**)(m2) - log(**write**)(m1) = β_{2}*(log(m2)
- log(m1))

It can be simplified to
log(**write**(m2)/**write**(m1)) = β_{2}*(log(m2/m1)), leading to

write(m2)/write(m1) = (m2/m1)^β_{2. }

This tells us that as long as the ratio of the
two math scores, m2/m1 stays the same, the expected ratio of the outcome variable,
**write**, stays the same. For example, we can say that for any 10% increase
in **math** score, the expected ratio of the two geometric means for writing
score will be 1.10^β_{2 }= 1.10^.4085369 = 1.0397057. In other
words, we expect about 4% increase in writing score when math score increases by
10%.

The content of this web site should not be construed as an endorsement of any particular web site, book, or software product by the University of California.