Transforming the Dependent Variable in Regression Models

MacKinnon, James G.; Magee, Lonnie

doi:10.2307/2526842

Cited by 222 publications

(97 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is suited for the transformation of dependent variables and allows consistent estimation of the regression equation (MacKinnon and Magee, 1990;Burbidge, Magee, and Robb, 1988).…”

Section: Test Of Hypothesismentioning

confidence: 99%

Trading Offshore: Evidence on Banks' Tax Avoidance

Langenmayr

Reiter

2017

SSRN Journal

View full text Add to dashboard Cite

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Little is known about how banks shift profits to low-tax countries. Because of their specific business model, banks use profit shifting channels different from those of other firms. We propose a novel and bank-specific method of profit shifting: the strategic relocation of proprietary trading to low-tax jurisdictions. Using regulatory data from the German central bank, we show that a one percentage point lower corporate tax rate increases banks' fixedincome trading assets by 4.0% and trading derivatives by 9.0%. This increase does not arise from a relocation of real activities (i.e. traders); instead, it stems from the relocation of book profits. Terms of use: Documents inJEL-Codes: H250, G210, F210.

show abstract

“…It is suited for the transformation of dependent variables and allows consistent estimation of the regression equation (MacKinnon and Magee, 1990;Burbidge, Magee, and Robb, 1988).…”

Section: Test Of Hypothesismentioning

confidence: 99%

Trading Offshore: Evidence on Banks' Tax Avoidance

Langenmayr

Reiter

2017

SSRN Journal

View full text Add to dashboard Cite

show abstract

“…However, log transformations do not allow zero or negative values, which are common in data concerning wealth. The IHS transformation addresses the issue of extreme values while allowing positive, zero, and negative values for our wealth variable (for more detail, see Burbidge, Magee, and Robb, 1988;MacKinnon and Magee, 1990;and Pence, 2006). 7 As outlined in detail by Burbidge, Magee, and Robb (1988), the IHS transformation has been used for data pertaining to wealth and health expenditures (Pence, 2006, andZhang et al, 2000), 8 specifically when variables take both extreme positive and negative values and, in many cases, the value of zero.…”

Section: Key Variablesmentioning

confidence: 99%

The Effects of Health and Wealth Shocks on Retirement Decisions

Conley¹,

Thompson²

2013

View full text Add to dashboard Cite

Retirement decisions both affect and are affected by health status. Health status, in turn, has been linked to net worth. And according to the life cycle model of savings, retirement has an important effect on net worth because retirees begin to expend their assets to maintain consumption once they leave the labor force. Given the multidirectionality of all these influences at this three-factor nexus, it has been difficult to separate the direct impacts of health and wealth on retirement from the reciprocal effect of retirement on health and wealth. This is our goal in the present article.Many studies have found that health shocks predict retirement decisions (see, e.g., Hagan, Jones, and Rice, 2009). For example, using fixed effects estimators and instrumenting subjective health by "health stock, " Disney, Emmerson, and Wakefield (2006) find that ill health strongly predicts early retirement among respondents older than 50 years of age in the British Household Panel Survey. Health limitations have also been shown to have a similar impact on early retirement decisions. However, the evidence does not completely support the claim that health shocks lead to exit from the labor force. For example, French (2005) finds that health is not among the more important determinants of job exit at older ages.Both health status and net worth can affect retirement decisions. In some cases, early retirement may be precipitated by a shock to an individual's health and/or economic status. The authors examine how health and wealth shocks affect retirement decisions. They use data from the Panel Study of Income Dynamics to estimate a first-differences model of health and wealth shocks on retirement over the course of the 2000s in the United States. Their results suggest that acute health shocks are associated with labor market exits for older American men but not women. These results appear particularly strong for blacks, whose labor force participation seems particularly sensitive to health status, which may be due to different occupations for blacks and whites. (

show abstract

“…The IHS transformation, first proposed by Johnson (1949), was introduced to econometrics by . 21 It has been used as an alternative to log transformations for the dependent variable MacKinnon and Magee 1990) and for explanatory variables (Layton 2001) with variables that can take on zero or negative values. Results can be interpreted in percentage terms, as in log models.…”

Section: Methodsmentioning

confidence: 99%