A concave log-like transformation allowing non-positive values

49 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Units of measurement and the inverse hyperbolic sine transformation (2020, methodology): “…and even negative values without any arbitrary manipulations of the original variable and it allows a similar interpretation of the regression results as the log transformation (see, e.g., Carroll et al., 2003; Carboni, 2012; Ravallion, 2017; Bahar and Rapoport, 2018; Bellemare and Wichman, 2019).”
Dealing with the Log of Zero in Regression Models (2019, unknown): “Then, the model is often estimated rather by adding a dummy variable to indicate such a treatment, but this alternative 3An extended concave version of this transformation has recently been provided by Ravallion (2017).”
Elasticities and the Inverse Hyperbolic Sine Transformation (2019, unknown): “…we mean by ‘large values of x’ in subsection ‘Caveats’. arcsinh–Linear Specification Now, consider an estimable equation of the form ỹ = + x + (4) 1 As Ravallion (2017) notes, in cases where a concave log-like transformation of negative data is necessary, the inverse hyperbolic sine…”