###  Check over the following for any errors before use

### Sample data from http://rcompanion.org/documents/RHandbookProgramEvaluation.pdf

if(!require(geoR)){install.packages("geoR")}

Turbidity = c(1.0, 1.2, 1.1, 1.1, 2.4, 2.2, 2.6, 4.1, 5.0, 10.0, 4.0, 4.1, 4.2, 4.1, 5.1, 4.5, 5.0, 15.2, 10.0, 20.0, 1.1, 1.1, 1.2, 1.6, 2.2, 3.0, 4.0, 10.5)

hist(Turbidity, col="gray")

library(geoR)

T.box = boxcoxfit(Turbidity, lambda2=TRUE)

T.box

lambda = T.box$lambda[1]

lambda2 = T.box$lambda[2]

if(lambda==0){T.norm = log(Turbidity + lambda2)}

if(lambda!=0){T.norm = ((Turbidity + lambda2) ^ lambda - 1) / lambda}

hist(T.norm, col="gray")

you may try AID package and change 0 to 0.1,  logarithm of zero is not defined 

hope this will work

I do not see a problem to receive negative values for your original 0's. The aim should be to get a more symmetric distribution, and depending on the distribution of your data, the 0 may get negative. Why you would not allow negative values after transforming them?

See the geoR package documentation for the definition, in the boxcox function.

https://cran.r-project.org/web/packages/geoR/geoR.pdf

Dear Pankaj Sir, thanks for your suggestion! I do not want to manipulate any of the data set moreover even if I change 0 to 0.1, the logarithm will result in negative values which I am trying to avoid.

Matt thanks for the suggestion I am not able to keep the negative values since the vector data that I have and I need to be transformed is for river flow which being negative will lead to problems for my further analysis. 

Salvatore, thanks for the link and the the documentation, I have already gone through it.

Attached is the formula from the documentation.  After you know lambda and lambda2, you just plug them in to transform your original data vector.

###  Check over the following for any errors before use

### Sample data from http://rcompanion.org/documents/RHandbookProgramEvaluation.pdf

if(!require(geoR)){install.packages("geoR")}

Turbidity = c(1.0, 1.2, 1.1, 1.1, 2.4, 2.2, 2.6, 4.1, 5.0, 10.0, 4.0, 4.1, 4.2, 4.1, 5.1, 4.5, 5.0, 15.2, 10.0, 20.0, 1.1, 1.1, 1.2, 1.6, 2.2, 3.0, 4.0, 10.5)

hist(Turbidity, col="gray")

library(geoR)

T.box = boxcoxfit(Turbidity, lambda2=TRUE)

T.box

lambda = T.box$lambda[1]

lambda2 = T.box$lambda[2]

if(lambda==0){T.norm = log(Turbidity + lambda2)}

if(lambda!=0){T.norm = ((Turbidity + lambda2) ^ lambda - 1) / lambda}

hist(T.norm, col="gray")

I know this is a little old but, I just wanted to add a big thanks to Salvatore for his example above! After hours of trying to work this out in R using the geoR vignette, I finally have my data BC transformed!

Dear Salvatore sir, from the code given by you I have installed geoR package but boxcoxfit function is not working. Iam stuggling with it. Iam using R version 3.2 I need box-cox power transmission parameters ,which I can use for negative values of data with 2 lambda parameters.Can anyone please help me out.

@ DurgaPriyanka, the code above works for me. Try restarting R or RStudio. And then make sure that the geoR package is installed. Does the example above not work for you?